A Study of Chinese Internet Homophones from the Systemic Functional Perspective

by

 Qian Tang

       

       

A thesis submitted in partial fulfillment of the requirements for the degree of    

Master of Arts              

Department of East Asian Studies

University of Alberta              

       

© Qian Tang, 2014

   

ii  

Abstract  

This thesis explores the phonetic features, morphological classification, and discourse function

of Chinese Internet homophones based on the Leiden Weibo Corpus (a corpus of Chinese micro-

blogging). The analyses show that more than half of the Chinese Internet homophones in the data

retain the same pronunciation as their base words, while the rest differ in initials, finals, or tones.

Based on their morphological structure, Chinese Internet homophones are categorized into

Chinese character homophones, alphabet homophones, numeric letter homophones, and blending

homophones. An examination of the data shows that Chinese character homophones and

alphabet homophones are used much more frequently than numeric letter homophones and

blending homophones. This study also explores the discourse function of Chinese Internet

homophones from the perspective of systemic functional grammar. Chinese Internet

homophones are used to achieve grammatical cohesion (conjunction, reference, ellipsis, and

substitution) and lexical cohesion (repetition, synonymy/antonymy, hyponymy/meronymy, and

collocation) in micro-blogging. Finally, the motivations for the use of Chinese Internet

homophones are discussed. This study contributes to our understanding of the dynamic

development and functions of homophonic puns in Chinese Internet language.

Key words: Chinese Internet homophones, cohesion, phonetic features, morphological

classification, mirco-blogging/Weibo

iii

Table of Contents

Abstract .......................................................................................................................................... ii  

List of Tables ................................................................................................................................. v  

Chapter 1 Introduction ................................................................................................................ 1  

1.1   Computer-Mediated Communication  ..............................................................................................................  1  

1.2 Internet Language and Weibo  .............................................................................................................................  2  

1.3 Chinese Internet Homophones  ............................................................................................................................  5  

Chapter 2 Theoretical and Methodological Background of the Study .................................... 9  

2.1 Theoretical Background  ......................................................................................................................................  9  2.1.1 Systemic Functional Grammar  ....................................................................................................................  9  2.1.2 Cohesion Theory  ........................................................................................................................................  11  

2.2 Methodological Background  .............................................................................................................................  20  

2.3 Previous Studies of Chinese Homophones  ........................................................................................................  22  

Chapter 3 Classification of Chinese Internet Homophones .................................................... 26  

3.1 An Overview of the Use of Chinese Internet Homophones and Their Base Words in the Data  ........................  26  

3.2 Phonetic Features of Different Types of Chinese Internet Homophones  ..........................................................  29  3.2.1 Types of Chinese Internet Homophones Based on Their Phonetic Features  .............................................  30  3.2.2 Narrow-Homophones  ................................................................................................................................  31  3.2.3 Broad-homophones  ....................................................................................................................................  34  3.2.4 Interim Summary  .......................................................................................................................................  44  

3.3 Morphological Features of Different Types of Chinese Internet Homophones  .................................................  44  3.3.1 Types of Chinese Internet Homophones Based on Their Morphological Features  ....................................  45  3.3.2 Chinese Character Homophones  ................................................................................................................  46  3.3.3 Alphabet Homophones  ..............................................................................................................................  51  3.3.4 Numeric Letter Homophones  .....................................................................................................................  53  3.3.5 Blending Homophones  ..............................................................................................................................  54  3.3.6 Interim Summary  .......................................................................................................................................  56  

3.4 Summary  ...........................................................................................................................................................  56  

Chapter 4 The Discourse Function of Chinese Internet Homophones .................................. 58  

4. 1 Grammatical Cohesion  ....................................................................................................................................  59  4.2.1 Reference  ...................................................................................................................................................  60  4.2.2 Ellipsis  .......................................................................................................................................................  65  4.2.3 Substitution  ................................................................................................................................................  71  4.2.4 Conjunction  ...............................................................................................................................................  74  

iv

4.3 Lexical Cohesion  ...............................................................................................................................................  76  4.3.1 Repetition  ..................................................................................................................................................  77  4.3.2 Synonymy/Antonymy  ................................................................................................................................  82  4.3.3 Hyponymy/Meronymy  ..............................................................................................................................  88  4.3.4 Collocation  ................................................................................................................................................  95  

4.3 Summary  ...........................................................................................................................................................  97  

Chapter 5 Conclusions ................................................................................................................ 98  

5.1 Summary of Findings  ........................................................................................................................................  98  

5.2 Discussion  .......................................................................................................................................................  101  5.2.1 Phonology of Mandarin Chinese  .............................................................................................................  102  5.2.2 Typing Method  ........................................................................................................................................  102  5.2.3 The Influence of Chinese Dialects  ...........................................................................................................  103  5.2.4 Internet Censorship by the Chinese Government  .....................................................................................  104  

5.3 Limitations of the Study  ...................................................................................................................................  105  

5.4 Implications of the Study  .................................................................................................................................  105  

Reference ................................................................................................................................... 107  

Appendix .................................................................................................................................... 118  

v

List of Tables  

Table 3.1.1 Overview of the Top 10 Most Frequently Used Homophones and Their Base Words in the Data………………………....………………………...................27 Table 3.2.1 The Four Basic Tones In The Scale Of Five Pitch Levels ……………………….....29

Table 3.2.2 Overview of Two Types of Chinese Internet Homophones ………………………..30

Table 3.3.1 Types of Homophones with Their Token Frequency and Distribution in the Data …………………………………………....46 Table 3.3.2 Morphological Structure of Chinese Character Homophones ……………….…......49

Table 3.3.3 Top 10 Most Frequently Used Chinese Character Homophones in the Data……….50

Table 3.3.4 Top 10 Most Frequently Used Forms of Alphabet Homophones …………………..52

Table 3.3.5 Most Frequently Used Forms of Numeric Letter Homophones………………….....54

Table 3.3.6 Top 10 Most Frequently Used Forms of Blending Homophones…………………...55

Table 4.1 Hu et al. (2005)’s Classification of Cohesion…………………………………………59

1

Chapter 1 Introduction

As the fastest-growing communication medium, the Internet has exerted a great influence on

language use. Internet language is a variant of social dialect. Its lexical system is tied to everyday

talk. Lexicon is the most active and fast-changing part of Internet language (Li, 2007, p.8). The

social factors of characterization of Chinese dialects and Chinese Government censorship

facilitate the emergence of the distinctive features of Chinese Internet language (Liu, 2011;

Inkster, 2010). This thesis focuses on the linguistic features and discourse function of Chinese

Internet homophones. In this chapter, the notions of computer-mediated communication, Internet

language and micro-blogging (Weibo in Chinese), and Chinese Internet homophones will be

introduced respectively. This chapter will provide the background information for the thesis.

1.1 Computer-Mediated Communication

 

Computer-mediated communication (CMC) is defined as the “communication that takes place

between human beings via the instrumentality of computers” (Herring, 1996). The Internet was

put into use in 19691 to share and exchange data. Through more than 40 years of development,

the Internet has become an indispensable communication medium for people of all walks of life.

From the 1970’s to the early 1990’s, the primary computer-mediated communication tool was e-

mail, with most of its users from academia, science, and information technology. In the 1990’s,

the Internet experienced a period of rapid growth which resulted in a significant increase in the

use of CMC among North American language educators (Thorne, 2008, p.2). E-mail remained

                                                                                                                         1 It was originally developed by the American Advanced Research Projects Agency. “The first group of networked computers [is] called ARPANET” (Thorne, 2008, p.1).

2

the primary communication tool, but Internet Relay Chat and instant messaging tools became

popular. Now, due to the development of communication technology and networks, the use of

cell phones and tablets is becoming more and more popular in communication. Newer

communication tools such as blogging and micro-blogging came into existence. Traditional

computer-mediated communication has been extended to digital-mediated communication.

The development of network technology along with the increase of Internet users facilitates the

use of Internet language. The following section will introduce Internet Language and a new

communication tool Weibo in China.

1.2 Internet Language and Weibo

Internet language is a type of language used in computer-mediated communication. It has been

widely used in BBS (bulletin board system), chat rooms, e-mail, on-line magazines and literature,

blogs, micro-blogs, etc. It is a social dialect in cyber space influenced by the Internet and its

users.

Yu (2001) argues that there are two types of Internet language utilized by Internet users. The first

type includes Internet terminologies and proper nouns. For example, 调制解调器

(tiao2zhi4jie3tiao2qi4 ‘modem’), 宽带 (kuan1dai4 ‘broadband’), and 传输控制协议

(chuan2shu1kong4zhi4xie2yi4 ‘Transmission Control Protocol/Internet Protocol’) belong to

Internet terminologies; 网虫 (wang3chong2 ‘Internet geek’), 黑客 (hei1ke4 ‘hacker’), and 虚拟

空间 (xu1ni3kong1jian1 ‘cyberspace’) are Internet proper nouns. The second type of Internet

3

language is the everyday language used by users on the Internet (Yu, 2001, p.2). The interaction

between Internet language and daily communication has been accelerated, due to the

advancement of new technology and the increase of Internet users. This thesis focuses on the

second type of Internet language. Among its linguistic features, the lexical features of Internet

language make it most distinctive from other types of social dialects.

 

Chinese scholars have studied Internet language as a variety of language, typically featuring

variability and individuation (Yan, 2005). Internet language shows uniqueness, simplicity,

causality, creativity, and vulgarity (Li, 2004). Although the social features of Internet users (e.g.

career, age, gender, education degree, etc.) are hidden, the style of Chinese Internet language

reveals that the majority of Internet users are students and young middle class with middle or

higher education. The communication between Internet users has formed a discourse community,

though it is not very stable. The content and pattern of Chinese Internet language has the

following features: irrationality, diversification, randomness, etc. (Xia, 2005; Liu & Qian, 2005;

Chen, 2008). The Internet in China also adapts to the needs of grassroots movements, and

Chinese Internet language has distinctive grassroots characteristics. It shows the creativeness of

grassroots movements. In addition, their on-line expressions reveal their opposition toward

authorities (Shi, 2009, p.2).

Crystal (2006) classified Internet language into six types based on different “Internet situations”:

the language of e-mail, the language of chat groups, the language of the virtual world, the

language of the World Wide Web, the language of instant messaging, and the language of

blogging. Since 2006, a new communication method has emerged known as micro-blogging.

4

Micro-blogging, a shortened broadcast medium, shares information to other subscribers in real

time through web and Wireless Application Protocol (WAP). Micro-blogs “allow users to

exchange small elements of content such as short sentences, individual images, or video links”

(Kaplan & Haenlei, 2011). The earliest and most popular micro-blogging service is Twitter. In

2009, one of the largest four portals in China, Sina, founded the earliest Chinese micro-blogging

service, Sina Weibo2. Following Sina Weibo there also appeared Sohu Weibo, 163 Weibo,

Tencent Weibo, and many others.

According to China Internet Network Information Center (CINIC), as of the end of 2012 the total

population of Internet users in China reached 564 million. Among the Internet users, there were

309 million micro-blogging users. 65% of the users used cell phone terminals to visit micro-

blogs3. The majority of micro-blog users were from Sina Weibo4.

Micro-blogging has become one of the most important ways of communication on the Internet,

influencing people’s lifestyles, how they transfer information and even public opinions, leading

China to a new “micro era” (Zhu et. al, 2013, p.204). People from all walks of life, whether

government officials or ordinary Internet users, can micro-blog information regarding politics

and government. The Chinese public security authorities have even opened a “Weibo 1105” on

the Internet where people can express “micro complaints”. This reflects the powerful force of

micro-blogging on public opinions.

                                                                                                                         2 Weibo (微博, wei1bo2) refers to Chinese micro-blogging services. 3 Source: http://companies.caixin.com/2013-01-16/100483154.html 4 Source: http://www.cq.xinhuanet.com/2013-02/21/c_114749460.htm 5 110 is an emergency phone number in China.

5

Compared with traditional communication media, micro-blogging is more open, more

convenient, and more interactive. First, its promptness ensures rapid dissemination of

information and timely feedback. Second, the constraint of background context is relatively weak.

The connection between preceding context and following context is poor. Third, there is

extensive use of homophones, abbreviations, other symbols, word conversions, and special

sentence patterns in micro-blogging communication (Xiang, 2012, p.13). As the extensive use of

homophones stands out in the language used in micro-blogging, this thesis will focus on Chinese

Internet homophones and their discourse function.

1.3 Chinese Internet Homophones

According to the Longman Dictionary of Applied Linguistics, homophones are defined as “words

which sound alike but are written differently and often have different meanings” (Platt, Richards,

& Weber, 1985, p.130). In this thesis, Chinese Internet homophone is defined as “the language

phenomenon resulting from intentional uses of words” which are “identical or similar in sound

for the purpose of achieving certain special effect” in Internet communication (Chen, 2011, p.iii).

The relation between the two identical-sounding words is called homonymy. Homophones can

be spelled the same (i.e. homographs and homonyms; e.g. ground ‘floor’ and ground ‘past tense

and past perfect tense of grind’); they can also be spelled differently (i.e. heterographs; e.g. dye

‘a colored substance’ and die ‘cease life’). Homophones may also apply to phrases (e.g. the

longer string 2EZ ‘too easy’), which includes two or more lexical items. In this thesis, the study

of homophones does not consider homophonic phrases.

6

Chinese homophones, xieyin ci, are considered to include the words that are identical or similar

in sound. The use of homophones always involves two words linked by phonetic similarity. In

Chinese, the original word is called the base word (Chen, 2011), and the derived word having the

same or close pronunciation as the base word is usually called homophone. The base word is the

sound source of the derived word. A word with the same sound as the base word is named a

narrow-homophone in this paper and a word with similar sound to the base word is named a

broad-homophone. For example, the word 脖友 (bo2you3, 脖 ‘neck’, 友 ‘friend’, 脖友 ‘friends

made through blogging’) is from the base word 博友 (bo2you3, 博 ‘blog’, 友 ‘friend’, 博友

‘friends made through blogging’). 脖友 is a narrow-homophone of 博友, as they have the same

pronunciation. 脖友 literally means ‘neck friend,’ but it is used to present ‘blogging friends’ with

a humorous effect created by the original meaning of 脖 (bo2 ‘neck’) that has the same sound as

博 (bo2 ‘blog’). In Chinese, the meaning of the homophonic word is related to the base word, as

the example of 博友 and 脖友 shows. A homophonic word can be used in communication with

or without the base word.

Chinese is full of homophones, which can be found in literary works and folk adages.

Homonymy is an important means to improve the effect of their expression. For example, in Cao

Xueqin’s Dream of the Red Chamber6, a character’s name 贾雨村 (Jia3 yu3cun1, 贾 ‘a Chinese

family name’, 雨 ‘rain’, 村 ‘village’, 贾雨村 ‘a person’s name’; 贾雨村 is a narrow-homophone

of 假语村 [言]7jia3yu3cun1(yan2); 假 ‘false’, 语 ‘language’, 村 ‘village’, 言 ‘speech’, 假语村

[言] ‘rumor’) implies that this person is not honest (Ning, 1975). Another character’s name, 甄

                                                                                                                         6 A masterpiece of Chinese literature, its first 80 sections are considered written by Mr. Cao Xueqin. 7 Only the pronunciation of the first three characters were used as the base for the creation of the homophone.

7

士隐 (Zhen1shi4yin3, 甄 ‘a Chinese family name’, 士 ‘bachelor ’, 隐 ‘hide’, 甄士隐 ‘a person’s

name’; 甄士隐 is a narrow-homophone of 真事隐 zhen1shi4yin3; 真 ‘true’, 事 ‘thing’, 隐 ‘hide’,

真事隐 ‘true things are hidden’) implies that he may be hiding some truth. In folk culture, there

are a large number of homophones known as “lucky words” (Fong 2000). For instance, in the

Chinese (lunar) New Year there must be fish on the table. This is because that 鱼 (yu2 ‘fish’) has

the same sound as 余 (yu2 ‘surplus’), expressing a wish to have more than needed in the New

Year.

Normally, Chinese homophony is introduced as a rhetorical device, using a word that has the

same or similar pronunciation as another word (i.e. base word) to express the meaning of the

base word and create certain effect (e.g. humour or satire). Chinese homophones are dynamic

uses of words that sound alike (Chen, 2011). When the word is in actual use, it can become a

homophonic pun of the base word. For example, 鱼 (yu2 ‘fish’) and 余 (yu2 ‘surplus’) are two

semantically unrelated words. Only under certain circumstances, such as at a Chinese New Year

table, 鱼 ‘fish’ connotes the meaning of 余 ‘surplus’.

As a result of the popularity of the Internet, homophones have been widely used in digital-

mediated communication through all types of tools. Chinese Internet homophones are

homophones used in the context of the Internet in China. The Internet seems to be a breeding

ground for Chinese Internet homophones. There is a reflexive relationship between Chinese

Internet homophones and language used in daily communication. For example, the Chinese

Internet homophone 神马 (shen2ma3, 神 ‘magic’, 马 ‘horse’, 神马 ‘what’; a homophone of 什

8

么 shen3me ‘what’) originates from an on-line novel and later became popular in daily

communication. Chinese Internet homophones can also come from daily communication, such as,

闹太套(nao4tai4tao4, 闹‘noisy’, 太 ‘too’, 套 ‘cover’, 闹太套 ‘a homophone of English not at

all’). This homophone came from a Chinese singer Huang Xiaoming. When he was singing the

song One Word One Dream, his pronunciation of not at all was very close to 闹太套. People

made fun of him, calling him 闹太套. Then the word 闹太套 became popular on the Internet.

The lexical system of Internet homophones is not very stable. Some of them might only be used

on the Internet for a short period of time and then disappear, while others could be long lasting

and gradually become part of the common lexicon.

Based on the background of computer-mediated communication and the popularity of a new

communication tool, micro-blogging/Weibo, Chinese Internet homophones reveal certain new

features, such as the new morphological structures and phonetic changes. In addition to Chinese

characters, alphabets and numeric letters can also be used to create homophones. The base words

can be from other languages, such as English and Japanese. In this thesis, I will explore the use

of Chinese Internet homophones and their base words in the data and phonetic and

morphological classification of Chinese Internet homophones (Chapter 3), and their discourse

function (Chapter 4). The next chapter will provide the theoretical and methodological

background of the study.

9

Chapter 2 Theoretical and Methodological Background of the Study

The past two decades have seen rapid development of systemic functional grammar (SFG) in

China and it has exerted great influence on Chinese linguistics (Zhang, D. J., 2006). However,

there is not much research on homophones in Chinese computer-mediated communication (CMC)

from the SFG perspective. In this study, I will use systemic functional grammar to examine

Chinese Internet homophones in a Chinese micro-blogging corpus. A brief discussion of the

theory of cohesion in SFG and the data used for this thesis will provide the background for the

analyses in the following two chapters. In Section 2.1, I will briefly introduce systemic

functional grammar. In Section 2.2, I will present the methodology of this study. In Section 2.3,

an overview of past studies of Chinese homophones will be provided.

2.1 Theoretical Background

The analysis of discourse function of Chinese Internet homophones in this thesis is based on Hu

et al. (2005)’s framework of cohesion developed from Halliday’s systemic functional grammar.

In this section, I will introduce the theory and development of SFG and the classification of

cohesion.

2.1.1 Systemic Functional Grammar

SFG is a theory of grammar, established by Michael Alexander Kirkwood Halliday (1985).

Guided by semantics, he proposed that language be considered as a regular resource that

describes system rather than structure. Systemic theory is “a theory of meaning as choice, by

10

which a language… is interpreted as networks of interlocking options” (Halliday, 1994, p.xiv).

Halliday describes SFG as a way in which “a language is interpreted as a system of meanings,

accompanied by forms through which the meaning can be realized” (1994, p.xiv). SFG is an

analysis of synthesis grammar based on the study of semantics through systems of lexical and

grammatical choices that serve functions within social environment (Halliday, 1985).

Halliday (1968) proposed that in all languages, the system has four functions: experiential,

logical, speech, and interpersonal. Halliday (1972) later revised the four functions: experiential

and logical functions were combined into ideational function (i.e. to understand the environment);

speech function was named as textual function; while the interpersonal function remained the

same (i.e. to act on the other in it). These three functions, known as the general functions of

language, were named metafunctions. Cohesion is included in textual function. Halliday &

Hasan (1976) stated that the fundamental difference of text and non-text is whether it has texture,

and whether this texture is formed by cohesive relation. Therefore, cohesion has become an

important area of research in discourse studies.

SFG is useful in explaining the functions of language. It enhances our understanding of the

functions of language in different types of discourse, including discourse in computer-mediated

communication. Halliday (1985 &1994) pointed out that the purpose of SFG is to provide a

theoretical framework for the analysis of both spoken and written discourse.

Halliday’s work is the most important development of the ‘London School’, which was founded

by J. R. Firth” (Butler, 1985, p. 1).8 Halliday’s experience studying in China with two well-

known Chinese linguists, Luo Changpei and Wang Li, also influenced the formation of his                                                                                                                          8 J. R. Firth is regarded as Halliday’s “main inspiration”. Firth studied language from a formal linguistics perspective, specializing in contextual theories of meaning and prosodic analysis.

11

linguistic theory. Halliday’s SFG “is in line with the tradition of Chinese linguistics, which is

‘meaning-oriented’” (Huang, 2002, p. 281 & 285). SFG has become one of the most influential

theories in Chinese linguistics (Zhang, D. J., 2006). Since 1980s, Hu started to introduce SFG in

China, and since then SFG has been the most popular and well-received linguistic theory in

China (Huang, 2011, p. 45). In China, SFG has been used to study different types of discourse,

such as spoken discourse (e.g. Zhang & Fang, 1996), written literature (e.g. Shi & Sun, 2001),

advertisement (e.g. Tang, 2005) and second language teaching (e.g. McDonald, 1992). Though

SFG has been used to examine the Chinese media discourse, it has not been widely used in

studying Chinese Internet language. The advantage of using Halliday’s theory in this thesis is

that it allows for the analysis of cohesion function of Chinese Internet homophones, and the

exploration of the semantic, social, and cultural factors underlying the use of homophones. SFG

will provide a new perspective to the study of Chinese Internet homophones.

2.1.2 Cohesion Theory

SFG provides the theoretical background for the study of cohesion function of Chinese

homophones in computer-mediated communication. Through years of development, the theory

of cohesion has been widely used to analyze written discourse. This section will discuss the

concept of cohesion, the development of the cohesion theory, and its classification.

12

2.1.2.1 Definition of Cohesion

Cohesion refers to “relations of meaning that exist within the text, and that define it as a text”

(Halliday & Hasan, 1976, p.4). It is “part of the system of a language” (Halliday & Hasan, 1976,

p.5). The construction of cohesion is stated as follows:

Cohesion occurs where the interpretation of some elements in the discourse is

dependent on that of another. The one presupposes the other, in the sense that it

cannot be effectively decoded except by recourse to it. When this happens, a

relation of cohesion is set up, and the two elements, the presupposing and the

presupposed, are thereby at least potentially integrated into text.

(Halliday & Hasan, 1976, p. 4)

Halliday & Hasan (1976) proposed cohesion functions of English discourse in detail while

pointing out the significance of cohesion in language systems in general. It is generally agreed

that Cohesion in English (Halliday & Hasan, 1976) marked the establishment of cohesion theory

and serves as the classical theory for cohesion studies. Before Halliday & Hasan (1976), studies

of cohesion in English were conducted by scholars such as Jakobson (1960), Halliday (1964),

and Hasan (1968). These studies provide the foundation of cohesion theory.

After the publication of Cohesion in English (1976), Halliday and Hasan, along with other

scholars, further developed the cohesion theory. For instance, in Language, Context, and Text:

Aspect of Language in a Social-semiotic Perspective (1985), Halliday and Hasan stated that

13

semantic relations are the basis of cohesion, and that there is a cohesive tie which connects

different textual units of a message. They pointed out that cohesion specifically refers to co-

referentiality, co-classification, and co-extension in internal components of discourse.

The following examples will demonstrate Halliday and Hasan (1976 & 1985)’s classification and

concept of cohesion.

Halliday and Hasan (1976) classified cohesion into reference, substitution, ellipsis, conjunction,

and lexical cohesion. Reference, substitution, ellipse, and conjunction are grammatical cohesions.

Lexical cohesion includes reiteration and collocation.

Reference occurs when a participant or circumstantial element serves as another element’s

reference point to make the information signalled for retrieval.

For example,

[2.1]

“I had a little nut tree

Nothing would it bear

But a sliver nutmeg

And a golden pear” (Halliday & Hasan, 1985, p. 73)

In this excerpt, it in the second line refers to a little nut tree in the first line. The information of it

is retrieved from the proceeding text little nut tree and the cohesion lies in the continuity of

14

reference. “This relationship of situational identity of reference is known as co-referentiality”

(Halliday & Hasan, 1985, p. 73).

Substitution occurs when one item is used to replace another.

For example,

[2.2]

“I play the cello. My husband does, too.” (Halliday & Hasan, 1985, p. 73)

In this excerpt, does in the second sentence substitutes play the cello, serving as a cohesive

device. Does and play the cello belong to an identical class, and this kind of relationship is also

referred as co-classification (Halliday & Hasan, 1985).

Ellipsis is “something left unsaid,” but implies “understood nevertheless,” where there is there is

a corresponding completed item which presents the unsaid information (Halliday & Hasan,1976,

p.142). For example:

[2.3]

“Joan brought some carnations, and Catherine some sweet peas.” (Halliday & Hasan, 1976,

p.143)

15

The two clauses in this text are in a parallel structure, and the second one can be interpreted as

Catherine bought some sweet peas, while in the text, bought is omitted. Bought serves as the

source of an elliptical item and provides a connection within the text.

Conjunction covers the use of adjunct-like elements to express certain meanings that presuppose

the presence of other components in the discourse. For example:

[2.4]

“They fought a battle. Afterwards, it snowed.” (Halliday & Hassan, 1976, p. 228)

Afterwards is acting as a cohesive agent to connect the two sentences by time sequence.

Lexical cohesion was treated as a parallel cohesion type as a subcategory of grammatical

cohesion. Cohesion is “the cohesive effect achieved by the selection of vocabulary” (Halliday &

Hasan, 1976, p. 274).

Lexical cohesion includes reiteration and collocation. They define reiteration as follows:

…[It] involves the repetition of a lexical item, at one end of the scale; the use of a

general word to refer back to a lexical item, at the other end of the scale; and a

number of things in between- the use of a synonym, near synonym, or

superordinate.

(Halliday & Hassan, 1976, p. 278)

16

Below is an example of reiteration:

[2.5]

“Accordingly…I took leave, and turned to the ascent of the peak. The climb is perfect easy…”

(Halliday & Hassan, 1976, p. 278)

In this example, the use of two synonyms, ascent and climb falls within the lexical cohesion. The

use of two synonyms repeated the meaning “uprise” and supplemented to the preceding content

to link the two sentences.

Collocation refers to those lexical items with a particular association (i.e. having co-occurrence

tendency) and are used together in the text.

Here is an example of collocation:

[2.6]

The king was in his counting-house, counting out his money,

The queen was in the parlour, eating bread and honey,

The maid was in the garden, hanging out the clothes.

Along came a blackbird and pecked off her nose. (Halliday & Hassan, 1976, p. 292)

Each pair of lexical items is in a recognizable lexico-semantic relation: king and queen; parlour

and garden; dish and eat. Their co-occurrence serves as collocational cohesion. Also, in this

17

example, King and queen are exclusive of each other but under the same superordinate monarch.

This type of relationship is named as co-extension (Halliday & Hasan, 1985).

In a later publication, Halliday (1985) re-categorized lexical cohesion into repetition, synonymy,

and collocation. 9 Halliday & Matthiessen (2004) proposed the classification of cohesion into

conjunction, reference, substitution & ellipsis, and lexical cohesion. Lexical cohesion includes

elaboration, extension, and collocation. Elaboration comprises of repetition, synonymy, and

hyponymy.

In sum, cohesion is regarded as a type of semantic relations (Halliday & Hasan, 1976; Halliday

& Hasan, 1985; Nunan, 1993; & Berry, 1994). Cohesion indicates semantic relation between one

discourse element and another, creating semantic consistency. Some researchers also consider

cohesion as a lexical or grammatical feature of discourse (e.g., Hoey, 1991; Huang, 1988 &

2001). It is an approach to connecting different textual units within discourse. Halliday & Hasan

(1976)’s reference, substitution, ellipsis, conjunction, and lexical cohesion are the specific

lexico-syntactic resources to achieve semantic cohesive relation. These textual items that create

cohesion in discourse are called cohesive devices.

 

2.1.2.2 The development of cohesion theory in China

 

The study of cohesion theory in China began in the 1980s. During this period, many scholars

published papers and monographs introducing cohesion from different perspectives (e.g. Wang,

1981; Xu, 1982; Hu, 1984; etc.).For example, Hu, Zhu, and Zhang (1989) published their book A                                                                                                                          9 There is no change to the classification in Halliday (1994).

18

Survey of Systemic Functional Grammar to introduce Halliday and Hasan’s cohesion theory in

China. In this book, they used examples from Chinese to explain the theory.

Since the 1990s, cohesion theory has been further developed. Hu (1993) discussed the

connection between intonation and cohesion. Hu (1994) further developed cohesion theory by

expanding the scope of structural cohesion. He argued that structural cohesion is the semantic

linkage that connects a word, phrase or clause whose meaning doesn’t clearly appear in a

structure. The semantic meaning of these words, phrases or clauses will be retrieved by

comparing the syntactical structure with the presupposed structure (Hu, 1994, p.69). Zhu (1995)

pointed out that understanding the context and the reader’s background knowledge contribute to

discourse cohesion, especially for discourse without explicit cohesive devices. However, in some

other studies, these factors tend to be ignored. Most studies only focus on cohesive devices.

Zhang (1999, 2000, & 2001) put forward a more detailed theoretical framework of cohesion.

Zhang (2001) argued that cohesion should be defined as “the textual meaning that organizes

meaning realized by formal features within language”, and also the relation “that connects the

text with the context of situation” (p. 23). For text-internal cohesion, Zhang (2001) proposed 11

general cohesive principles. Aside from the development from Halliday & Hasan (1976), the

principles include mental distance (the distance of cohesion) and natural language’s order. His

contributions to cohesion theory also include introducing the cross-category cohesive mechanism,

exophora cohesion, recessive cohesion, and other types of cohesion.

19

2.1.2.3 Hu et al. (2005)’s Classification of Cohesion

Based on Chinese discourse, Hu et al. (2005) classified cohesion into grammatical cohesion and

lexical cohesion. Grammatical cohesion includes conjunction, reference, ellipsis, and substitution.

Lexical cohesion includes repetition, synonymy/antonymy, hyponymy/meronymy, and

collocation. Hu et al. (2005)’s definition of cohesion along with conjunction, reference, ellipsis,

substitution, and lexical cohesion are from Halliday & Hasan (1976). However, Hu et al. (2005)

treated grammatical cohesion and lexical cohesion as parallel subcategories of cohesion, which is

different from Halliday & Hasan (1976), Halliday (1985), and Halliday & Matthiessen (2004).

This thesis will adopt Hu et al. (2005)’s classification.

The reasons for using Hu et al. (2005)’s classification of cohesion are three-fold. First, Hu et al.

(2005)’s categorization keeps the traditional classification of cohesion based on Halliday &

Hasan (1976): grammatical cohesion and lexical cohesion. However, they put grammatical and

lexical cohesion at the same level. Halliday & Hasan (1976), Halliday (1985), and Halliday &

Matthiessen (2004) consider lexical cohesion as a subcategory of grammatical cohesion.  In other

words, grammatical and lexical cohesion are at different levels. However, there is no clear

evidence that they belong to different levels. Second, based on the principal of economy, the less

the better. Hu et al. (2005)’s classification is the most economical, while it also includes all types

of cohesion in discourse. Third, Hu et al. (2005)’s classification of cohesion is based on and is

adapted to Chinese discourse. It is the most suitable for the analysis of discourse function of

Chinese Internet homophones. Thus, Hu et al. (2005)’s classification is adopted in this study.  

20

2.2 Methodological Background

This section will introduce the data used in this thesis. The data are from the Leiden Weibo

Corpus (LWC, http://lwc.daanvanesch.nl/). The corpus contains messages from a Chinese micro-

blogging service Sina Weibo. The LWC is created by Daan van Esch. It contains 5.1 million

messages from Sina Weibo that were posted between January 8th, 2012 and January 30th, 2012.

All messages in this database were processed using the Penn Chinese Treebank standard for

Chinese word segmentation and part-of-speech tagging from the Stanford Natural Language

Processing Group. These messages come from real communication rather than being constructed.

Each message in the LWC has a unique message ID. Each message contains the user’s screen

name, location, and gender, provided by the users of Sina Weibo. At the home page

http://lwc.daanvanesch.nl/, messages can be searched by lexical items. From each message, the

searching result also shows the number of occurrences, frequency, and frequency rank of a

specific target word. Because the LWC relies heavily on computer natural-language processing,

the best ones can reach approximately 95% accuracy rates (Daan van Esch, 2012). Considering

the large quantity of the data, “these accuracy rates are similar to or better than the inter-

annotator accuracy rate for human annotators” (Daan van Esch, 2012). Because of this, the LWC

is an ideal corpus to study Chinese Weibo language.

In order to examine a large number of messages within a reasonable time, this thesis uses a

dataset of 18,791,530 words taken from the Leiden Weibo Corpus based on a random sampling

method. As homophonic pans (homophones and their base words) are hard to identify in the

corpus without reading each message from the dataset, an initial word list of Chinese internet

homophones was constructed based on the Green Paper on Language Situation in China

21

(GPLSC, 2006-2013), Chinese Internet language dictionaries, Sina Weibo Niandu Pandian (Sina

Weibo Annual Report, 2011-2012). The GPLSC has been published annually by the Ministry of

Education and National Language Committee of China since 2006. GPLSCs are official reports

on contemporary Chinese language. These reports include statistical data based on their nation-

wide database, lists of new words with their meanings, and commonly-used words from the

Internet. These sources list all new Internet words or/and commonly-used Internet words each

year. The Chinese Internet language dictionaries that I have consulted for this study include New

Internet Dictionary and Chinese Internet Language Dictionary. I manually identified Internet

homophones from all the word lists from the aforementioned sources. The GPLSCs provide

reliable information on Chinese Internet language. The dictionaries and the Sina Weibo Annual

Reports provide information from non-governmental perspectives. Thus, arguably the initial list

constructed from these different sources is a relatively accurate representation of contemporary

Chinese Internet homophones. The initial list of 150 different homophones was imported to

CasualConc 1.9.710 to count each homophone’s token frequency from the dataset that was taken

from Leiden Weibo Corpus. Ranked by token frequency, a final word list that contains 117

Chinese Weibo homophones based on the dataset was built11.

In the analysis of phonetic features of Chinese Internet homophones, I used IPA to notate the

Chinese words (with the exception of Section 3.1, Chinese words in other parts of this thesis are

notated with Chinese Pinyin12). In the analysis of the morphological classification of Chinese

Internet homophones, an overview of each type of homophone with their frequency and                                                                                                                          10 CasualConc 1.9.7 is a concordance software program which can generate concordance lines, word clusters, collocation analysis, and word counts. It works on the English language, UTF-8 text, etc. Details can be found at http://casualconc.blogspot.ca/. 11 There are 33 Chinese Internet homophones from the initial word list that didn’t appear in the dataset. 12 Chinese Pinyin refers to the Romanized phonetic system which transcribes the pronunciation of Chinese characters using Latin script. It was officially published by the Chinese government. This system is widely used for teaching Mandarin Chinese, as a method of inputting Chinese characters, and so on.

22

percentage is presented. Under each category, the most frequently used homophones from the

final word list are provided with Chinese Pinyin notation, the corresponding base words, and the

percentage of each homophone’s occurrences in the total occurrences of the homophones and its

base word(s).

The excerpts containing homophones in Chapter 4 are all from the final word list, and each

excerpt comes with a unique message ID which can be looked up at the LWC

(http://lwc.daanvanesch.nl/). In the data, some messages contain emoticons (i.e. pictorial

representations of facial and emotional expressions) and hyperlinks. I chose to only look at the

text of each message. This thesis keeps the original version of the messages which include

mistakes made by micro-blog users in words, sentences, and punctuation. For each excerpt, a

gross translation is provided: the first line shows the original data in Chinese characters; the

second line is the Pinyin notation; the third line shows the English meaning of each word; and

the fourth line provides the English translation for the entire line.

2.3 Previous Studies of Chinese Homophones

 

Since the introduction of systemic functional grammar in China, the theory of cohesion has been

largely applied to study different types of Chinese discourse. However, the theory of cohesion is

not widely used to study Chinese homophones in computer-mediated communication (CMC).

The study of Chinese Internet homophones will provide new insight into cohesion in Chinese

CMC. This section will offer a brief overview of past studies of Chinese homophones.

23

The discourse function of Chinese homophones is examined in different types of discourse. For

example, Zheng (2007) explored discourse function and register features of cell phone text

messages. Zheng (2007) pointed out that homophones can be agents of cohesion in text messages

and they create intertextuality. Intertextuality refers to the connection and interaction between

two or more texts. In Zheng (2007), it refers to the connection and interaction of a chain of cell

phone text messages. By exploring the textual mechanism of Chinese homophones in the

discourse of text messaging, Li (2009) studied the phonetic, morphological, discourse, and

semantic features of Chinese homophones. She also explored lexical cohesion and grammatical

cohesion created by homophones. She stated that there are mainly two types of grammatical

cohesion used in text messages: ellipsis and reference. Wang (2005) studied means of cohesion

based on grammar and sound, and pointed out that the correlation of semantic meanings in

discourse is essential. The use of homophones in allegorical sayings,13 poems, and numeric letter

messages can achieve textual cohesion. Chen (2007) examined Chinese cross talks14 from the

data Xiangsheng mingzuo yu xingshang (Masterpiece of cross talks and appreciation). He argued

that the use of homophones can realize cohesion in cross talks and achieve humorous effect.

Wang (2010) studied cohesion of Chinese antithetical couplets (i.e., two parts of a pair of lines of

poetry written on a pair of scrolls) from the perspective of lexis and sound. Wang (2010) argued

that homophony can be a device to realize textual cohesion. By exploring the phonetic features

of homophonic puns, Bo (2007) analyzed cohesion function of homophonic puns in Chinese

classical literature. Bo (2007) put forward that the use of homophonic puns depends on the user’s

understanding of the context and the purpose of expression. Zhang (2007) studied the lexical and

                                                                                                                         13Allegorical sayings, i.e. xiehouyu in Chinese, which in the Oxford Language Dictionaries Online (http://www.oxfordlanguagedictionaries.com) is defined as “a two-part allegorical saying, of which the first part, always stated, is descriptive, while the second part, sometimes unstated, clinches the point”. 14 Chinese cross talks, i.e., comic dialogues, usually involving two persons.

24

grammatical features of new Internet words. He pointed out that new Internet homophones can

be created by borrowing phonetic sounds from English, Chinese dialects, Chinese characters,

numeric letters, and Chinese acronyms.

In addition, Chinese homophones have been studied from other perspectives, such as culture (e.g.

Zhao, 1987 & Fong, 2000), rhetoric (e.g. Zhang, 2005; Lu & Li, 2003 ), and psychology (e.g.

Siok & Fletcher, 2001). From a culture perspective, Zhao (1987) examined the relation of

homophones and Chinese culture, and pointed out that Chinese homophones are related to Han

peoples’ cultural tradition and the ethnic group’s psychology.15 Zhang (2005) explored Internet

language from a perspective of rhetoric. He indicated that homophony, a type of rhetoric of the

Internet language, can be more expressive for the purpose of realizing humorous and pleasant

effects. Lu & Li (2013) explored the rhetorical effect of Chinese homophonic puns by means of

semantic field theory, and explicated the factors that affect rhetorical effects. From a psychology

perspective, Siok & Fletcher (2001) studied phonological awareness and visual-orthographic

skills in Chinese reading comprehension. They found that among their students, the success of

reading for students from Grade 2, 3, and 5 is affected by the ability to distinguish homophonic

characters.

Through the overview of the past studies of Chinese homophones, we can see that homophones

are widely used in all types of discourse to achieve textual cohesion. Homophony is common in

cell phone text messages, allegorical saying, poems, cross talks, antithetical couplets, written

                                                                                                                         15 Han is the majority ethnic group in mainland China. Similarly, Fong (2000) explored Chinese New Year “luck talk” that consists of speech acts related to luck, and discussed it with the principle of Yin and Yang and the theory of relative relativism. She stated that New Year food and the use of “luck talk” to counteract bad luck are homophones of the lucky words in Chinese, symbolizing luckiness, abundance, wealth, etc.

25

literature, etc. However, these studies did not provide clear classification of cohesion used in

their analyses. For example, there is no clear statement about the classification of cohesion in Li

(2009). Wang (2010) treated homophony as a type of cohesion. Chen (2007) studied

homophones under the category of structural cohesion. However, their analyses of cohesion in

discourse do not seem to be thorough enough. In addition, most of the papers did not indicate

how the data were collected (e.g. Li, 2009), which makes their analyses less convincing. Thus, in

this study, I use the data from a large established corpus based on a Chinese micro-blogging

service. I will systematically explore the discourse function of Chinese Internet homophones in

this corpus, by adopting Hu et al. (2005)’s classification of cohesion.

In the next chapter, I will explore the use of Chinese Internet homophones and their base word

(Section 3.1), the classification of Chinese Internet homophones based on their phonetic features

(Section 3.2), and the morphological structures (Section 3.3). Chapter 4 discusses the discourse

function of Chinese Internet homophones.

 

 

26

Chapter 3 Classification of Chinese Internet Homophones

The use of homophones is a prominent feature of Chinese Internet discourse. When a

homophone is formed, its pronunciation and its morphological structure may differ from its base

word. This chapter will explore the classifications of Chinese Internet homophones based on

their phonetic and morphological features. In Section 3.1, an overview of the use of Chinese

Internet homophones and their base words in the data will be provided. In Section 3.2, I will

examine the classification of Chinese homophones based on their phonetic features. The

phonetic features of narrow-homophones and broad-homophones will also be explored. In

Section 3.3, I will investigate the morphological classification of Chinese Internet homophones.

This classification includes Chinese character homophones, alphabet homophones, numeric letter

homophones, and blending homophones. In the second section of this chapter, the details of the

phonetic features of Chinese Internet homophones will be discussed. These homophones are

presented with IPA phonetic notation, while in other parts of this chapter, they are presented with

Chinese Pinyin.

3.1 An Overview of the Use of Chinese Internet Homophones and Their Base Words in the Data

As mentioned in Chapter 1, Chinese homophones are always derived from their base words, and

they are linked by their phonetic similarities. When faced with the two kinds of written forms, i.e.

the homophones and the base words, some Internet users choose to use homophones, while

others still use the original forms. The following table provides an overview of the use of the top

10 most frequently used homophones and their base words in the randomly sampled dataset.

27

Table 3.1.1 Overview of the Top 10 Most Frequently Used Homophones and Their Base Words in the Data

                                                                                                                         16  The percentage is calculated based on the formula: [occurrences of the homophone/ (occurrences of the homophone + occurrences of its base word(s) )]*100%.  

Homophone (H)

Homopho-nic meaning

Occur-rences of H

H in the total

occurrences (%)16

Base word (B)

Occur-rences of B

B in the total occur -rences (%)

SB sha2bi1

idiot, asshole

17, 469 92.46% 傻逼 sha2bi1 ‘idiot

/asshole’

1, 424 7.54%

神马 shen3ma3

‘magic horse’

what

9, 139 11.97% 什么 shen3me1

‘what’

67, 219 88.03%

围脖 wei2bo2

‘wrap neck’

mirco-blog 5, 152 30.98% 微博 wei2bo2 ‘micro

blog’

11, 480 69.02%

尼玛 ni2ma3

‘priestess agate’

fuck off

4, 534 99.85% 你妈 ni3ma1 ‘your

mom’

7 0.15%

粉丝 fen3si1

‘vermicelli’

fans

4, 495 92.30% fans ‘enthusiastic followers of a

sport, celebrity, etc.’

375 7.70%

RP ren2pin3

‘personality’

personality, generalized

for luck

3, 706 67.60% 人品 ren2pin3

‘personality’

1,776 32.40%

88 ba1ba1

‘bye-bye’

bye-bye 3, 334 65.17% 拜拜 bai4bai4 ‘bye-

bye’

1,782 34.83%

表 biao3

‘watch’

don’t 3, 081 91.48% 不要 bu2yao4 ‘not

want’

287 8.52%

28

As is shown in the Table 3.1.1, among the top 10 most frequently used homophones, 7 of them

are used more frequently than their base words in the data. For example, the Chinese Internet

users never use the base word 囧 (ancient Chinese) jiong3 ‘bright’, but always use its

homophone 囧 jiong3 ‘glooming or embarrassing’. The homophones 尼玛 ni2ma3 ‘fuck off’, SB

sha2bi1 ‘idiot, asshole’, 粉丝 fen3si1 ‘fans’, and 表 biao3 ‘don’t’ are also used much more

frequently (99.85%, 92.46%, 92.30%, and 91.48%) than their base words. However, 肿么

zhong3me1 ‘how’, 神马 shen2ma3 ‘what’, and 围脖 wei2bo2 ‘micro-blog’ seem to be used less

frequently (7.14%, 11.97%, and 30.98%) than their base words. Table 3.1.1 shows that among

the top 10 most frequently used homophones, the homophones are generally used rather

frequently compared to their base words. However, Chinese Internet homophones are not very

stable in the Chinese lexical system. Some of the Internet homophones may replace the base

words and come into the Chinese common lexicon, while others may disappear.

The following sections will explore the phonetic features and morphological features of Chinese

Internet homophones.

肿么 zhong3me1

‘swelling/swell me

(particle)

how

2, 837 7.14% 怎么 zen3me1 ‘how’

36, 874 92.86%

囧 jiong3

‘expressing the look of gloom or

embarrass’

glooming, embarrass

-ing

1, 716 100% 囧(ancient Chinese)

jiong3 ‘bright’

0 0%

29

3.2 Phonetic Features of Different Types of Chinese Internet Homophones

Chinese language has a writing system which is character-based with each graph corresponding

to one syllable. In Old Chinese, the majority lexical items are monosyllabic (i.e., each word is

one syllable in length). Through the development of disyllabification from Old Chinese to

Modern Chinese, there has been an increase of disyllabic words or multi-syllabic words in the

lexicon system (Pi, 1992, p. 56). Mandarin Chinese syllables are composed of initials, finals, and

tones. 17 In the Mandarin phonological system, there are 2 initials, 39 finals, and 5 tones (4 basic

and 1 neutral).

Given the small number of initials and finals, the Mandarin phonological rules, and the

convention of spelling,18 there can be only 398 types of initial-final combinations, and

17 In Mandarin syllable structure, syllables are divided into initials, finals and tones. Initials are the optional consonants that can occupy the first position (Sun, 2006, p.35). The other part of the syllable is called a final. For example, ta55 [ta55] ‘he’ is constituted by the initial t [t], the final a [a] and a flat tone (the Mandarin tones will be explained at Table 3.1.1); deng55 [təŋ55] ‘light’ has the initial d [t], the final eng [əŋ] and a flat tone. All Mandarin initials are consonants, but the consonant ng cannot function as an initial. Essentially, a Mandarin final consists of vowel(s) and optional nasals (please see the graph of Mandarin syllabus structure in 3.2.3.1).

Chinese tones change the pitch of syllables and make them lexically contrastive. The 4 basic tones in Mandarin Chinese are the high-level tone (the first tone), the high-rising tone (the second tone), the low-rising tone (the third tone) and the falling tone (the fourth tone). There is also a short and weak neutral tone, in addition to the 4 basic tones. As the neutral tone is short and weak and the pitch level is decided by the previous syllable, it is not described in this paper. The scale of five pitch levels of Mandrin tones is set forth by Yuen Ren Chao (1930). The pitch values of the 4 basic tones in Mandarin are 55, 35, 214 and 51 respectively as shown in table 3.2.1 with examples. Table 3.2.1 The Four Basic Tones in the Scale of Five Pitch Levels

Pitch 1st Tone High-level tone 55/HH

2nd Tone High-rising tone 35/MH

3rd Tone Low-rising tone 214/LH

4th Tone Falling tone 51/HL

5/ High4/High 3/Mid 2/Low 1/Low

Example

[ma 55] ‘mather’

[ma 35] ‘hemp’

[ma 214] ‘horse’

[ma 51] ‘scold’

18 In Mandarin Chinese pinyin spelling rules, some initials and finals cannot be combined. For example, the initial f [f] cannot be spelled with i [i] or any final that begins with i [i]. Some initials and finals can be spelled together according to the conventional

30

approximately 1,200 distinct tonal syllable types (Suen, 1979). However, according to Modern

Chinese Frequency Dictionary (1986), there are 4,574 characters in common usage. The limited

number of syllables in Mandarin is conducive to the generation of homophones. The

pronunciation of Chinese homophones is drawn from Mandarin, Chinese dialects, and

occasionally other languages. The next section will offer a classification of Chinese homophones

based on their phonetic features.

3.2.1 Types of Chinese Internet Homophones Based on Their Phonetic Features

Through homophonization, Chinese homophones may have the same or similar pronunciation as

their base words. Based on Chinese linguistic traditions, Chinese homophones are classified into

two types: homophones with the same sound as their base words and homophones with a similar

sound to their base words (Chen, 2011). In this chapter, homophones with the same

pronunciation as their base words are named narrow-homophones, whereas homophones with a

similar pronunciation to their base words are called broad-homophones. Within the category of

narrow-homophones, there are homographic homophones (homophones with the same spelling

as their base words) and heterographic homophones (homophones that have a different spelling

than their base words). All broad-homophones are heterographic homophones.

In the dataset, there are 117 homophones. Among these homophones, 64 are narrow-

homophones and 53 are broad-homophones. Table 3.2.2 provides an overview of the two types

of homophones.

                                                                                                                                                                                                                                                                                                                                                                                                       phonology rules, but there are no such corresponding words in Mandrin. For example, the initial f [f] can be spelled with the final ai [ai] theoretically, but it doesn’t occur in Mandarin.

31

Table 3.2.2 Overview of Two Types of Chinese Internet homophones

Type Numbers %

Narrow-homophones 64 54.70%

Broad-homophones 53 45.30%

Total 117 100%

Table 3.2.2 shows that, in the data, Chinese Internet homophones with the same pronunciation as

base words (narrow-homophones) are slightly more frequent than those with similar

pronunciation (broad-homophones). The following sections will elaborate on the phonetic

features of narrow-homophones and broad-homophones in Chinese Internet discourse. The

presentation of Mandarin Chinese initials and finals is transcribed using the International

Phonetic Alphabet (IPA). The 4 basic tones  are marked based on Chao (1930)’s scale of five

pitch levels (see Footnote 17).

3.2.2 Narrow-Homophones  

As mentioned earlier, narrow-homophones are homophones which have the same pronunciation

as their base words. Within narrow-homophones there are homographic homophones

(homophones with the same spelling as their base words) and heterographic homophones

(homophones that have a different spelling than their base words.  The following sections will

explore the two types of narrow-homophones: homographic narrow-homophones and

heterographic narrow-homophones.

32

3.2.2.1 Homographic Narrow-homophones

Homographic narrow-homophones possess the same orthographic form and pronunciation as

their base words. For instance, the homophone 囧 [tɕyŋ] 214 has the same appearance and

pronunciation as its base word, the ancient written form 囧 [tɕyŋ214] ‘bright’. Based on the look of

this character, the homophone 囧 has been used to express the look of gloom or embarrassment

by internet users. However, the meaning of the homophone 囧 [tɕyŋ] 214 ‘gloom or

embarrassment’ is completely different from the meaning of the ancient word 囧 [tɕyŋ214] ‘bright’.

Another example is the homophone 槑 [mei  35] ‘very silly’ which has the same written form as

base word−the ancient character 槑 [mei  35] ‘plum’ (equivalent to the simplified character 梅 [mei  

35] ‘plum’). However, the meaning of the homophone 槑 comes from the meaning of each of the

two components 呆 [tai55] ‘silly’ rather than from the meaning of the ancient character 槑.

3.2.2.2 Heterographic Narrow-homophones

Heterographic narrow-homophones refer to words having the same pronunciation but different

orthographic forms from base words. Based on the data, the homophonic pronunciation of the

syllable(s) of heterographic homophones can be found in Chinese characters, alphabets, and

numeric letters. The following section will elaborate on the phonetic features of character,

alphabet, and numeric letter heterographic narrow-homophones.

33

First, character heterographic narrow-homophones are heterographic homophones with the same

sounds as Chinese characters. In the dataset, the heterographic homophone 杯具 (杯[pei  55] ‘cup’,

具[tɕy51] ‘tool’, 杯具 ‘cup’) which has an original meaning of ‘cup’ and a homophonic meaning

of ‘tragedy’, is read as [pei  55 tɕy51], the same pronunciation as its base word 悲剧 [pei  55   tɕy51]

‘tragedy’, but they are written in different Chinese characters. The homophone 脖友 [po53  iou214]

‘friends made through blogging’ (脖 [po53] ‘neck’, 友 [iou214] ‘friend’) and the base word 博友

[po53  iou214] ‘friends made through blogging’ are a pair of homophonic puns. The character 脖

[po35] ‘neck’ in 脖友 has the same pronunciation as the character 博 [po35] ‘blog’ in the base

word 博友. Although they have the same pronunciation, character heterographic narrow-

homophones have totally or partially different written forms from base words.

Alphabet heterographic narrow-homophones refer to the alphabet(s) or letters in heterographic

narrow-homophones that represents the same sound as the base word. Most alphabets in

homophones represent the same pronunciation of the Chinese characters found in the base words.

In the data, there are two specific types of alphabet heterographic narrow-homophones. The first

type is that alphabets are used to represent the sound of the base word. The second type is that

alphabets are used as acronyms to represent the sound of the base word. As a full Chinese Pinyin

form, KAKA [kha55 kha55] has the same sound as its base word 咔咔 [kha55 kha55], an

onomatopoeic word for laughter. DING [t  iŋ214] also has the same pronunciation as its base word

顶[t  iŋ214] ‘to support or to push up the ranking of a posting’. Other examples show that acronyms

represent the full pronunciation of the base words, such as LX [lou35  ɕia51] ( base word 楼下[lou35  

ɕia51] ‘downstairs’). On BBS (bulletin board system), it usually refers to the comment posted

34

after the original post or the previous comment in a series of discussion. The alphabetic letters in

ZF [tʂ  əŋ51 fu214] (base word 政府 [tʂ  əŋ51 fu214] ‘government’) and 傻B [ʂa214  pi55] (base word 傻

逼 [ʂa214  pi55] ‘fool’) also represent the full pronunciation of their base words.

Finally, numeric letter heterographic narrow-homophones are homophones in which the numeric

letters represent the same pronunciation as the base words. 19 For example, 2货 [ɚ51xuo51] ‘idiot’

is the heterographic narrow-homophones of 二货 [ɚ51xuo51] ‘idiot’, and 2B [ɚ51pi55 ] ‘stupid’ is

the heterographic narrwo-homophones of 二逼 [ɚ51pi55 ] ‘stupid’.

3.2.3 Broad-homophones

Broad-homophones are heterographic homophones which have a similar pronunciation to their

base words. Broad-homophones can be further divided into three types: Chinese character,

alphabet, and numeric letter broad-homophones. The following sections will provide detailed

information on each type of broad-homophones.

3.2.3.1 Chinese Character Broad-homophones

In this section, as the discussion involves Mandarin Chinese syllable structure, an example with a

Mandarin initial and final structure diagram is presented as follows.20 Mandarin finals are

composed of optional medials, main vowels, and optional syllabic terminals. Medials, also

                                                                                                                         19 Conventional written forms in Arabic numbers are excluded, such as 1 in 第 1 名 ‘rank on the first place’, and 98 in 98 年 ‘the year of 1998’, etc. 20 For the definition of Mandarin initials and finals, please see footnote 16.

35

known as on-glides, are optional sounds in the first position of a final.21 The main vowel is the

essential vowel in a final.22 A terminal, also known as off-glide, is an optional sound in the last

position of a final.23 For example, in the syllable tian thiɛn ‘sky’, t is a initial, ian is a final;

among the segments of the final, i serves as a medial, a a main vowel, and n a terminal.

t i a n

C V V C initial final medial main terminal vowel Of the three components of a Mandarin syllable (i.e., initials, finals and tones), at least two are

the same as their base words in Chinese character broad-homophones (Li, 2007). There are three

types of the different phonetic features compared with the base words. The first one is that the

initial and final of a homophone are the same as its base word, while the tone differs. This type is

very common. The second type is that a homophone retains the same tone, and the same main

vowel and terminal (of the final) as its base word. This is also very common. The third type is

that a homophone may have a similar initial and tone to its base word, but a different final. It is

because that a final is the most important part in a Chinese syllable and it is unlikely to change

(Zhou, 2005). This type of phonetic feature is relatively rare. Homophones generally have the

same main vowels and terminals as their base words. The following sections will summarize the

phonetic features of initials, finals, and tones of Chinese character broad-homophones compared

with their base words. In addition, the phonetic features of Chinese transliteration homophones

21 Only [i], [u],[y] in Mandarin Chinese may be used as medials (e.g., the segment [i] in the syllable tian ‘sky’). 22 Main vowels in Mandarin are [a], [ə], [ɛ], [o], [i], [y] (e.g., the segment [a] in the syllable tian ‘sky’). 23 Only [i], [u], [o], [n], [ŋ] may be used as Mandarin terminals (e.g., the segment [n] in the syllable tian ‘sky’).

C

itial

V

fin

V V V

d

V C

nal

m

V

na

V

ain termai

36

will be addressed at the end, as their base words are from other languages, which is out of the

scope of the Mandarin syllable system.

3.2.3.1.1 Initials of Chinese Internet Homophone

First, summarizing the phonetic features of homophone initials involves explaining both their

places and manners of articulation. Homophones tend to have the same manner, but different

places of articulation from their base words. The most common cases are outlined below.

(1) Post-alveolar affricates/fricatives in homophones corresponding to dental

affricates/fricatives in their base words and vice-versa

It is common that post-alveolar affricates/fricatives in homophones correspond to dental

affricates/fricatives in their base words. For example, the homophone 肿么 [tʂuŋ214 me51] ‘how’

and its base word 怎么[tsəәn214 me51] ‘ how’ have the same manner of articulation (i.e., they share

the manner of affrication, but they differ in articulators). The base sound of [ts] is a dental

affricate, while the homophonic sound [tʂ] is a post-alveolar affricate. The same feature of

initials is also found in the homophone 妹纸 [mei55  tʂ  ʅ214] ‘sister’ or ‘girl’, and its base word 妹

子 [mei55  tsɿ214] ‘sister’ or ‘girl’,  as well as the homophone  汉纸 [xan55  tʂ  ʅ214  ] ‘man’ or ‘true man’

and its base word 汉子[xan55  tsɿ214]  ‘man’ or ‘true man’, etc. In addition, a dental sound in

homophones may also correspond to a post-alveolar sound in the base words. This

correspondence is also seen in the dataset. For example, the dental fricative [s] in the homophone

桑心 [sɑŋ55ɕ  in55] ‘sad’ corresponds to the post-alveolar fricative [ʂ] in the base word 伤心 [ ʂɑŋ55ɕ  

in55] ‘sad’.

37

Here, the initials of Mandarin homophones are usually related to those of their base words.

Specifically, when they have the same manner of articulation, the place of articular of the

homophone initials is usually different from that of the base words. This type of difference may

be influenced by local dialects, where there may not be a differentiation between sounds that

have the same place of articulation, or the same manner of articulation. For example, [n] and [l],

[f] and [x] are not differentiated in Xiang dialect.

(2) A labiodental fricative [f]in homophones corresponding to a velar fricative [x] in their base

words and vice-versa

The sounds [f] and [x] have the same manner of articulation(i.e., fricative), but different places

of articulation. [f] is a labiodental fricative, but [x] is a velar fricative. For instance, the

labiodental fricative [f] in the homophone 稀饭 [ɕi55  fan51] ‘porridge’ corresponds to the velar

fricative [x] in the base word 喜欢 [  ɕi214    xuan55] ‘like’. However, the velar fricative [x] in the

homophone 灰常 [xueɪ55  tʂhɑŋ35] ‘very’ can also correspond to the labiodental [f] in the base word

非常 [  fei55  tʂhɑŋ35] ‘very’.

(3) A nasal [n] in homophones corresponding to a lateral [l] in their base words

The sounds [l] and [n] have the same place of articulation; both are alveolars. But they have

different manners of articulation. [l] is an alveolar lateral, whereas [n] is an alveolar nasal. For

example, the initials [n] of the first two Chinese characters of the homophone 内牛满面 [nei51  

niou35  man214  miɛn51] ‘a face bathed in tears’ (内 ‘inner’, 牛 ‘ox’, 满‘full,’ 面 ‘face’) correspond

38

to [l] in the base word 泪流满面 [lei51  liou35  man214  miɛn51]  ‘a face bathed in tears’ (泪 ‘tear’, 流

‘flow’).

(4) A  lateral l [l] or a zero-initial in homophones corresponding to a post-alveolar approximant

r[ʐ]  in their base words  

A  lateral l [l] or a zero-initial in homophones corresponds to a post-alveolar approximant  r[ʐ] in

the base words. For example, the lateral [l] in the homophone 伦家 [luən  35  tɕia55] ‘I’ corresponds

to the post-alveolar approximant [ʐ] in the base word 人家 [ʐuən  35  tɕia55] ‘I’. Yet the zero-initial

in the homophone 男淫 [nan35  in35] ‘man’ corresponds to the sound [ʐ] in the base word 男人

[nan35  ʐ  en35] ‘man’.24

3.2.3.1.2 Finals of Chinese Internet Homophone

Mandarin finals include medials, main vowels, and syllabic terminals.25 Comparing the finals of

homophones and those of their base words, there seems to be more regular patterns. The medials

of homophones tend to be unrounded, while those of the base words are rounded.26 The phonetic

features of main vowels involves the degree of openness (vowel height), the location of the

active part of the tongue (vowel backness), and lip rounding (vowel rounding) (Lin, 2007, p.63).

In the data, generally, the main vowels of homophones tend to be higher and unrounded. In

addition, for the phoneme /e/ in the base words, a high front unrounded [i] in homophones

corresponds to a mid-central unrounded vowel [ə] in the base words, and a low central vowel [a]

                                                                                                                         24 The homophone 淫[in35] only has the main vowel and the terminal. 25 For the diagram of Mandarin Chinese final structure, please see Page 35. 26 Glides refer to transitional sounds.

39

in homophones corresponds to a mid-high front unrounded vowel [e] in the base words. The

phonetic features of homophones also involve the addition or deletion of main vowels compared

to their base words. As for terminals, alveolar nasal and velar nasal can interchange, but there

doesn’t seem to be regular rules in place. Some typical cases of finals in the data are outlined

below.

(1) A high front unrounded medial [i] in homophones corresponding to a high front rounded

medial [y] in their base words

Through homophonization, a high front unrounded medial [i] in homophones corresponds to a

high front rounded medial [y] in the base words. For example, [i] in the homophone 童鞋[thuŋ35  ɕ  

iɛ  35] ‘classmate’ corresponds to the medial [y] in the base word 同学[thuŋ35    ɕyɛ  35]  ‘classmate’.

(2) The phoneme /e/ in the base words

In the data, the phoneme /e/ in the base words can be realized as a mid-central unrounded vowel

[ə] and a mid-high front unrounded vowel [e].

A high front unrounded [i] in homophones corresponding to a mid-central unrounded vowel [ə]

in the base words. For example, [i] in the homophone 男淫[nan  35  in35  ] ‘man’ corresponds to the

main vowel [ə] in the base word男人[nan  35  ʐ  ən35] ‘man’.

40

A low central vowel [a] in homophones corresponds to a mid-high front unrounded vowel [e] in

the base words, and vice-versa. For instance, [a] in the homophone 神马 [ʂəәn35 ma 214] ‘what’

corresponds to [e] in the base word 什么[ʂəәn 21 me55] ‘what’, whereas [e] in the homophone

特么[the51  me55] ‘damn it’ or ‘fuck’ corresponds to [a] in the base word 他妈 [tha  55 ma  55]

‘damn it’ or ‘fuck’. In addition, a high front unrounded vowel [i] in homophones corresponds to

[e] in the base words. For example, [i] in the homophone 虾米 [ɕia  55  mi214] ‘what’ corresponds

to [e] in the base word 什么[ʂəәn 21 me55] ‘what’. A back rounded diphthong [uo] corresponds to

[e] in the base words as well. For example, [uo] in the homophone 帅锅 [ʂuai  51  kuo55]

‘handsome man’ corresponds to [e] in the base word 帅哥[ʂuai  51  ke55] ‘handsome man’.

(3) Addition or deletion of main vowels

When considering the phonetic features of homophones’ finals, there can be addition or deletion

of vowels. Addition of vowels can make a diphthong in homophones correspond to a

monophthong in the base words. For example, the diphthong [uə] in the homophone 伦家 [l  uən  

35  tɕia55] ‘I’ corresponds to the single vowel [ə] in the base word 人家[ʐ  ən35  tɕia55] ‘I’ (the initial

[l] corresponds to [ʐ] ). The diphthong [ou] in the homophone 偶棉 [ou214 miæn35] ‘we’

corresponds to the main vowel-monophthong [o] in the base word 我们[uo55 m  ən  35]27 ‘we’. The

deletion of vowels may make a monophthong in homophones correspond to a diphthong in the

base words. For example, the monophthong [i] in the homophone 米国[mi214 kuo25] ‘USA’

corresponds to the diphthong [ei] in the base word 美国[mei214 kuo35 ] ‘USA’.

                                                                                                                         27 The sound [u] in 我 [uo55] is the medial, and it disappeared in the homophone.

41

(4) A velar nasal terminal [ŋ] in homophones corresponding to an alveolar nasal terminal [n] in

their base words and vice-versa

A velar nasal terminal [ŋ] in homophones corresponds to an alveolar nasal terminal [n] in their

base words, and vice-versa. For instance, the velar nasal terminal [ŋ]  in  the homophone  肿么

[tsuŋ  214  me55] ‘how’ corresponds to the alveolar nasal terminal [n] in the base word 怎么 [tsən  

214  me55] ‘how’. The alveolar nasal terminal [n] in the homophone 盆友 [ph  ən53  iou  214] ‘friend’

corresponds to the velar nasal terminal [ŋ]  in the base word  朋友 [phəŋ53  iou  214] ‘friend’.

3.2.3.1.3 Tones of Chinese Internet Homophones

In the data, there is no fixed pattern related to the tones of Chinese Internet homophones

compared with their base words. The tones appear to be somewhat random, mainly due to the

way that words are typed on a computer keyboard. Chinese is a tonal language, and keyboard

input cannot reflect tones. The frequency of the words popped up on the screen is based on

common lexicon. Often instead of choosing the correct word, Internet users pick a homophone,

which appears first among the many alternative forms with the same initial and final on the

computer keyboard. Sometimes, they are new lexical items, such as 矮油‘oh’, ‘wow’, ‘my God’

[ai214  iou35] (base word 哎呦[ai55  io55] ‘oh’, ‘wow’, ‘my God’), or 怪蜀黍‘weird uncle’[k  uai51  

ʂu214  ʂu35] (base word 怪叔叔[k  uai51  ʂu55  ʂu55] ‘weird uncle’). In other cases, homophones are

based on pre-existing words in Chinese, such as 斑竹[pan55  tʂu35]’mottled bamboo’ (base word

版主[pan214  tʂu214] ‘board moderator’), 鸭梨[ia55  li35] ‘pear’ (base word 压力[ia55  li51] ‘pressure’),

42

and 河蟹[xe35  ɕiɛ51] ‘river crab’(base word 和谐[xe35  ɕiɛ51] ‘harmonious’, a political concept of

the Chinese government). However, these cases are not commonly seen in the current data.

In addition, as homophones may also have base words originating from other languages, the

Chinese transliteration homophones undergo certain sound changes from the pronunciation

found in the original language. As these transliteration words use Chinese pronunciation, the

forms are not entirely the same as those of their base words. This type of homophone often

borrows similar sounds from Chinese to express certain meanings. Some changes that can occur

during transliteration include addition of tone, and addition or subtraction of syllables. Some

transliteration words have a significant sound change, while others undergo slight sound changes.

For example, 粉丝 [fən 214 sɿ55], transliterated from the English word fans [fæns], is changed

from one syllable to two syllables in addition to tone and vowel changes. The added vowel [ɿ]

adds one syllable to the homophone. 卢瑟 [lu35  se51] is mapped from the English pronunciation

loser [ˈluzɚ]. Although there is no change to the number of syllables, there is an addition of

Chinese tones (this automatically leads to a change in pitch). The latter syllable [zɚ] in the base

word loser is changed to [se] in the homophone. Usually the changes involve close consonants or

vowels changes and tone addition. In the database, there are only minor sound changes in

Chinese transliteration words.

3.2.3.2 Alphabets and Numeric Letters

Most of the alphabet homophones retain the pronunciation of Chinese characters. The letters in

alphabet homophones are usually acronyms of the Pinyin form of the base words and therefore

43

represent the pronunciation of their base words. For example, the homophone ZF [tʂ  əŋ51 fu214]

‘government’ comes from the base word 政府 [tʂ  əŋ51 fu214] ‘government’. However, there are

some exceptions that are more complex and more random. For example, the word string 3Q

referring to the English word ‘thank you’ comprises of the Chinese pronunciation of the numeric

letter 3 [san55] and the English pronunciation of Q [kju:]. As for V5 [vei55  u214] (base word 威武

wei1wu3[uei55  u214] ‘mighty’ ) , the letter V ( a labiodental fricative) replaces the first letter W of

威’s Pinyin form.28 It is likely that this spelling and sound change is influenced by an existing

Chinese dialect. The sound change from [u] to [v] can be found in the Chinese Ningxia dialect

and the Gulin dialect (Huang & Liao, 2008, p.46).

As for numeric letters, a single numeric letter can represent different homophonic sounds. For

instance, 3 [san55] in 3166 [san55  i55  liou51  liou51] (base word Japanese さよなら[ˌsaɪəәˈnɑːrəә],

‘goodbye’) is a homophonic sound to さ[sa], 3 in 1314 [i55  san55  i55  sɿ51] (base word 一生一世[i51  

ʂəŋ55  i35  ʂʅ51] ‘in all one's life’) is a homophonic sound of 生 and 3 in 3Q [san55  kju:]( base word

thank you) is a homophonic sound of than [θæŋ]. These examples show that numeric letter

homophones have certain variability. In addition, the pronunciation of numeric letters in some

dialects co-exists with that of Mandarin in the Chinese Internet homophones. For example in

1314 (一生一世) and B4 [pi214  sɿ4] (base word 鄙视 [pi214  ʂʅ51] ‘to despise or disdain’), 4 [sɿ51]

presents the homophonic sound of 世 [ʂʅ51] and 视[ʂʅ51], which originates from some southern

dialects. In these southern dialects, there is no distinction between dental and post-alveolar

initials. So in this case, people use the dental [s] to replace the post-alveolar [ʂ].

                                                                                                                         28 In Mandarin Pinyin spelling, when the final is a single u, w will be added in front of u; if the final begins with u, then u will be changed to w.

44

3.2.4 Interim Summary

The special phonetic features make homophony an interesting linguistic phenomenon in Chinese.

Chinese Internet homophones are formed with identical pronunciation (narrow-homophones) or

similar pronunciation to their base words (broad-homophones). The data show that the number of

narrow-homophones is slightly higher than that of broad-homophones. In this section, I have

explored the phonetic features of narrow-homophones and broad-homophones. Narrow-

homophones can have the same spelling as their base words (homographic) or different spelling

from their base words (heterographic). Most of the alphabet homophones are narrow-

homophones. Broad-homophones, which are heterographic homophones, can be divided into

homophones with similar sounds to Chinese characters, alphabets, and numeric letters.

Specifically, the analysis of the phonetic features of Chinese character broad-homophones is

based on the changes to initials, finals, and tones from their base words. To sum up, more than

half of the Chinese Internet homophones have the same pronunciation as their base words, while

comparing with the base words, the initials of homophones tend to differ in the place of

articulation, while the manner of articulation remains the same. Vowels in homophones tend to

be unrounded and are higher than the vowels in the base words. Chinese dialects and even

foreign languages can influence the phonetic features of Chinese Internet homophones.

3.3 Morphological Features of Different Types of Chinese Internet Homophones

 

Based on their morphological features, homophones in Chinese Internet discourse can be

(Chinese) character homophones, alphabet homophones, numeric letter homophones, or a blend

of characters, alphabets and numeric letters. However, after a homophone is formed, its

45

morphological type may be changed. This section will offer a classification of homophones

based on their morphological structures.

3.3.1 Types of Chinese Internet Homophones Based on Their Morphological Features  

Traditionally, Chinese morphological classification is solely based on Chinese characters.

However, in addition to Chinese characters, Chinese Internet homophones are comprised of

alphabets and numeric letters. Based on their morphological structure, Zhou (2008) categorized

Chinese Internet lexicon into four types: Chinese characters, alphabets, numeric letters, and

blends ( p. 434). This study adopts Zhou’s categorization.

Among the 117 Chinese Internet homophones identified in the data, there are 55 character

homophones (47%), 46 alphabet homophones (39.32%), 4 numeric letter homophones (3.42%),

and 12 blending homophones (10.26%). Here I also provide an overview of the occurrence and

distribution of each type of homophone in the data. The occurrences of character homophones,

alphabet homophones, numeric letter homophones, and blending homophones are shown in

Table 3.3.1 below.

46

Table 3.3.1 Types of Homophones With Their Token Frequency and Distribution29 in the Data

Type Token Frequency %

Chinese character homophones 41,641 55.08%

Alphabet homophones 29,223 38.66%

Numeric letter homophones 3,763 4.98%

Blending homophones 965 1.28% Total 75,592 100%

Table 3.3.1 shows that the Chinese character homophones occurred the most frequently

(55.08%) followed by the alphabet homophones (38.66%), the numeric letter homophones

(4.98%) and the blending homophones (1.28%). It can be seen that character homophones and

letter homophones are used significantly more frequently than numeric letter homophones and

blending homophones. Character homophones are the most frequently used type. This result is

consistent with Zhou (2008a)’s finding that Chinese character homophones are the most

frequently used in Internet communication, compared to other types of homophones. The

following sections will elaborate on the morphological features of each type of homophones.

3.3.2 Chinese Character Homophones

A Chinese word, ci, is defined as the smallest meaningful language unit which can be used

independently (Hu, 1992, p.230). Morphemes are the minimal units of meaning. A Chinese word

                                                                                                                         29 The distribution is calculated based on the formula:  𝐹! =

!!!∗ 100%, where Fi is the target i’s frequency, ni is the occurrences of

target i, and N is the total occurrences of all targets.

47

consists of one or more morphemes. Based on the number of morphemes that the words have,

they are classified into single-morphemic words and multi-morphemic words. In Chinese, the

number of morphemes and that of characters (one character corresponds to one syllable) are not

in a corresponding one-to-one relationship. In this section, the examples in Chinese will be noted

with Chinese Pinyin.

Firstly, a single-morphemic word only has one morpheme, including monosyllabic and

polysyllabic single-morphemic words. For instance, 败 bai4 ‘defeat’ (base word English buy) is

a single-morphemic word, and it only has one syllable. However, 特么 te4me1 ‘damn

it/fuck’(特,‘special’, 么, ‘me (particle)’) with the corresponding base word 他妈 ta1ma1 ‘damn

it/ fuck’ (他 ‘his’, 妈 ‘mother’) is a single-morphemic word composed of two

syllables/characters. Single-morphemic homophones with two or more syllables are considered

to be polysyllabic single-morphemic words, such as 特么.

Secondly, multi-morphemic words consisting of more than one morpheme include compounds,

reduplications, and derivations (Gu, Pan, & Liu, 2001, p.10). Most words in modern Chinese are

compounds. Compounding is the most productive method of word-formation in Chinese (Huang

& Liao, 2003; Hu, 1992; etc.). For instance, 杯具 bei1ju4 ‘cup /tragedy’ (base word 悲剧

bei1ju4 ‘tragedy’) has two morphemes, 杯 bei1 ‘cup’ and 具 ju4 ‘tool’. The morpheme 洗 xi3

‘wash’ and another morpheme 具 ju4 ‘tool’ can be combined together, forming the compound

word 洗具 xi3ju4 ‘cleaning set/comedy’ (base word 喜剧 xi3ju4 ‘comedy’).

48

“Reduplication is a morphological process in which a root or stem or part of it is repeated”

(Hartmann & Stork, 1972, p. 193). Some scholars categorize reduplicated words into multi-

morphemic words (Huang & Liao, 2002), while others consider them as single-morphemic

words (Sun, 2006). In this study, reduplicated words are treated as multi-morphemic

words.  There are no reduplicated homophones found in the data.30

Derivation is the formation of a new word from another word or stem (Hartmann & Stork, 1972,

p. 62). For instance, the Chinese suffix 子 zi3 ‘zi’ is usually attached to morphemes and denotes

small or round things, kinship terms or persons. 筒子 tong3zi ‘thick tube-shaped thing/ comrade’

(base word 同志 tong2zhi4 ‘comrade’) is constituted by the root morpheme 筒 and the suffix 子.

A summary of the morphological structure of Chinese Internet homophones in the data is shown

in Table 3.3.2.

                                                                                                                         30 They are only examples of reduplicated words found in the base words. If we look at the example of the morpheme 妹 mei4 ‘sister’, it can be repeated to create the base word 妹妹 mei4mei ‘sister’ (homophone 美眉 mei3mei2, 美‘beautiful’, 眉 ‘eyebrow’, 美眉 ‘beautiful girls’).

49

Table 3.3.2 Morphological Structure of Chinese Character Homophones

Table 3.3.3 provides an overview of the top 10 most frequently used character homophones and their morphological structure found the data.

Single-morphemic words

Monosyllabic single-morphemic words

Polysyllabic single-morphemic words

Multi-morphemic words

Compounds

Reduplications

Derivations

 50

Table 3.3.3 Top 10 Most Frequently Used Chinese Character Homophones

Homophone   Homophonic meaning Base word   Morphological structure   Occurrences  

神马 shen3ma3 ‘magic horse’  

what 什么 shen3me ‘what’  

polysyllabic single-morphemic word  

9,139  

围脖 wei2bo2 ‘wrap neck’  

mirco-blog 微博 wei2bo2 ‘micro blog’  

compound word   5,152  

尼玛 ni2ma3 ‘priestess agate’

fuck off

你妈 ni3ma1 ‘your mom’  

polysyllabic single-morphemic word  

4,534  

粉丝 fen3si1 ‘vermicelli’

fans

fans Enthusiastic followers of

a sport, celebrity, etc

polysyllabic single-morphemic word  

4,495  

表 biao3 ‘watch’  

don’t 不要 bu2yao4 ‘not want’

monosyllabic single-morphemic word  

3,081  

肿么 zhong3me1 ‘swelling/swell

me (particle)  

how

怎么 zen3me1 ‘how’  

polysyllabic single-morphemic word  

2,837  

囧 jiong3 ‘expressing the look

of gloom or embarrass’

glooming, embarrassing

囧 (ancient Chinese) jiong3 ‘bright’

 

monosyllabic single-morphemic word  

1,716  

虾米 xia1mi3 ‘shrimp rice or little

shrimp’

what 什么 shen3me1 ‘what’  

compound word   1,575  

擦 ca1 ‘wipe’  

fuck 操 cao4 ‘fuck’  

monosyllabic single-morphemic word  

961  

特么 te4me1 ‘special me

(particle)’  

damn it/ fuck 他妈 ta1ma1 ‘his mother’  

polysyllabic single-morphemic word  

927  

 

51

Within the category of character homophones, single-morphemic words are used more frequently

than multi-morphemic words. This is because homophones are created based on pronunciation,

and the newly created words (homophones) do not tend to reflect the number of morphemes in

the base word. For example, in some southern dialects, the multi-morphemic word 不要 (bu2

yao4 ‘don’t’) sounds like a monosyllabic word biao3. As a result, some internet users started to

type the single-morphemic word 表 (biao3 ‘watch’) as a homophone of 不要.

3.3.3 Alphabet Homophones

Alphabet homophones are written in alphabetic letters, which derive from the Pinyin form of

each Chinese base word. Some alphabet homophones are written in Pinyin or Pinyin with some

variations. For example, Ding is a homophone of the base word 顶 (ding3 ‘to support’ or ‘to

push up the ranking of a posting’), while HOHO is a homophonic pun for the base word 吼吼

(hou3hou, ‘an indicator of emotion representing happiness or excitement’). Acronym

homophones are a major type within the category of alphabet homophones. For example, 人品

ren2pin3 ‘luck’ is the base word for the homophone RP. When Internet users see PR, they read it

as ren2pin3, rather than pronouncing the alphabets.

Table 3.3.4 provides a summary of the top 10 most frequently used alphabet homophones in the

data. As alphabet homophones are not case-sensitive, lower case, upper case, and combination of

the two are counted in the data. For instance PR, Pr, and pr are all homophones of 人品

52

ren2pin3 ‘luck’, but in the table only the most frequent form, PR, is used to represent all the

spellings.

Table 3.3.4 Top 10 Most Frequently Used Forms of Alphabet Homophones

Homophone Homophonic meaning Base word

Occurrences    

SB

idiot, asshole

傻逼 sha2bi1 ‘idiot /asshole’

17,469

RP personality, generalized for luck

人品 ren2pin3 ‘personality’

3,706

MM

sister, generalized for girls 妹妹

mei4mei ‘sister’

1,703

girls, especially beautiful girls

美眉 mei3mei2 ‘girls’

YY unrealistic imagination

意淫 yi4yin2 ‘unrealistic

imagination’

832

ZZ to forward others' posting

转载 zhuan3zai3 ‘to forward

others' posting’

460

LG husband 老公

lao3gong1 ‘husband’ 452

BT abnormal 变态

bian4tai4 ‘abnormal’ 447

PP

butt 屁屁

pi4pi ‘butt’

384

picture 片片

pian1pian ‘picture’

HOHO

indicator of emotion, showing happiness or

excitement 吼吼

hou3hou ‘modal particle’

350

Generally, alphabets are easy to type. This appears to make alphabet homophones one of the

most frequently used types in the data. However, the pronunciation of alphabet homophones is

53

different from English abbreviations.31 For instance, WTO stands for World Trade Organization

in English and is pronounced as the three alphabets, which is different from what we saw above.

However, polysemous alphabet homophones (alphabet homophones with two or more different

meanings) may cause ambiguity. For most of the alphabet homophones in Chinese, each of them

corresponds to one base word. But some alphabet homophones correspond to two or more base

words. For instance, PP is a homophone of 屁屁 (pi4pi ‘butt’) and 片片 (pian1pian ‘picture’). In

these particular cases, the reader needs to rely on the context to decide which the base word is.

3.3.4 Numeric Letter Homophones

Numeric letter homophones consist of Arabic numbers based on two types of pronunciations.

Numeric letters in the first type present the same or close pronunciation in Chinese. As to the

second type, numeric letters present the same or close pronunciation of other languages, such as

Japanese. For example, Internet users use the numeric letter homophone 88 (ba1ba1 ‘bye-bye’;

base word 拜拜 bai4bai4 ‘bye-bye’), which comes from the English word byebye, when they say

goodbye to each other over the Internet. The frequency of numeric letter homophones is

relatively low. The total occurrences in the data are 3,763 (4.98 % of all Chinese Internet

homophones). 32 Table 3.3.5 shows the numeric letter homophones found in the data.

                                                                                                                         31 In addition to their pronunciation being different than English abbreviations, Chinese alphabet letter homophones are non-case sensitive, while English abbreviations need to be capitalized. 32 In the current study, phrases with numeric letters are excluded, such as 520 wu3er4ling2 ‘I love you’ (base phrase 我爱你

wo3ai4ni3 ‘I love you’), 530 wu3san1ling2 ‘I miss you’ (base phrase 我想你 wo3xiang3ni3 ‘I miss you’), 7456 qi4si4wu3liu4 ‘piss me off’ (base phrase 气死我了 qi4si3wo3le1 ‘piss me off’). Their base expressions are longer than lexical units, which are beyond the scope of this study.

54

Table 3.3.5 Most Frequently Used Forms of Numeric Letter Homophones

Homophone Homophonic meaning

Base word Occurrences

88 ba1ba1

bye-bye 拜拜 bai4bai4 ‘bye-bye’

3,334

555 wu3wu3wu3

crying 呜呜呜 wu1wu1wu ‘imitative word representing crying’

352

1314 yi1san1yi1si4

in all one's life

一生一世 yi4sheng2yi2shi4 ‘in all one's life’

66

3166 san1yi1liu4liu4

bye-bye さようなら (Japanese)

Sayonara ‘bye-bye’

11

Sometimes, one numeric letter can represent different characters and thus has different

pronunciations. As is shown in table 3.3.5, 3 in 1314 represents the characters 生 sheng1,

whereas 3 in 3166 represents the syllable さ sa in Japanese.

3.3.5 Blending Homophones  

Blending homophones are a mix of characters, alphabets and/or numeric letters. The frequency

of blending homophones is the lowest among the four morphological types, totaling 965 in the

corpus (1.28% of the total occurrences). Table 3.3.6 lists top 10 most frequently used forms of

blending homophones.

55

Table 3.3.6 Top 10 Most Frequently Used Forms of Blending Homophones

Homophone Homophonic meaning Base word Occurrences

V5 vei1wu3 mighty and domineering

威武 wei1wu3 ‘power military’

442

2B er4bi1 stupid

二逼 er4bi1 ‘two female- genitalia’

412

B4 bi3si4 to despise 鄙视

bi3shi4 ‘despise look-on’ 63

3Q san1Q thank you thank you 21

傻 B sha3bi1 fool

傻逼 sha3bi1 ‘stupid female- genitalia’

14

小 P 孩 xiao3pi4hai2 young man (pejorative)

小屁孩 xiao3pi4hai2 ‘little butt child’

5

2 货 er4huo4 idiot 二货

er4huo4 ‘two goods’ 4

P 民 pi4min2 shitizen 屁民

pi4min2 ‘fart citizen’ 1

装 B zhuang1bi1 cocky

装逼 zhuang1bi1 ‘pretend female-genitalia’

1

牛 B niu2bi1 freaking awesome

牛逼 niu2bi1 ‘ox female- genitalia’

1

Here we can see that blending homophones are constituted by a blend of characters and numeric

letters, such as 2 货 er4huo4 ‘idiot’ (base word 二货 er4huo4 二 ‘two’, 货‘goods’, 二货

‘idiot’).We also find combinations of characters and alphabetic letters, such as 装 B zhuang1bi1

‘cocky’ (base word 装逼 zhuang1bi1 装 ‘pretend’, 逼 ‘female-genitalia’, 装逼 ‘cocky’).

Blending homophones can also be formed by the combination of alphabetic  letters and numeric

56

letters. For example, B4 bi3si4 ‘despise’, ‘look-on’, or ‘to despise’ is based on the Chinese

pronunciation of the base word 鄙视 bi3shi4 ‘despise’, ‘look-on’, or ‘to despise’. Unlike the

homophones in Section 3.2.3, where alphabets are based on the Chinese pronunciation, here we

see another type. Some alphabets in blending homophones have a similar pronunciation to

English. For instance, 3Q san1q ‘thank you’ is based on the Chinese pronunciation of the

numeric letter 3 san1 and the English pronunciation of the letter q.

 3.3.6 Interim Summary  

The morphological classification of Chinese homophones from the data reveals the diversity in

the methods used to form Chinese Internet homophones. Internet homophones are composed of

not only Chinese characters, but also alphabets and numeric letters. Alphabet homophones,

numeric letter homophones, and blending homophones are not formed through the traditional

methods of word-formation in Chinese. This shows the creativity and vitality of Chinese

Internet language.

From the statistical overview of each morphological type of Chinese Internet homophones in the

data, the most frequently used homophones are character homophones, with a frequency of

55.08%. The second most frequently used homophones are alphabet homophones, with a

frequency of 38.66%.

3.4 Summary  

In this chapter, the phonetic features and morphological structures of Chinese Internet  

57

homophones have been studied. Section 3.1 provides an overview of the use of the Chinese

Internet homophones and their base words. Among the Top 10 most frequently used

homophones in the data, the homophones are generally used rather frequently compared to their

base words. Section 3.2 examines the phonetic features of Chinese Internet homophones and

classifies them into narrow-homophones and broad-homophones. More than half of the

homophones have the same pronunciation as their base words. In broad-homophones, the initials

of homophones tend to differ from their base words, when the place of articulation or the manner

of articulation remains the same. Vowels in broad-homophones tend to be unrounded and higher

than those of the base words. Section 3.3 explores the morphological features of Chinese Internet

homophones. They are classified into four types, Chinese character homophones, alphabet

homophones, numeric letter homophones, and blending homophones. In the data, character

homophones and alphabet homophones are used much more frequently than numeric letter

homophones and blending homophones.  

 

 

58  

 

Chapter 4 The Discourse Function of Chinese Internet Homophones

This chapter will analyze the discourse function of Chinese Internet homophones based on the

classification of cohesion proposed by Hu et al. (2005) within the theoretical framework of

systemic functional grammar (SFG) (Halliday & Hassan, 1976; Halliday, 1985). Cohesion is the

connection within a text that makes it a semantic unit (Halliday & Hasan, 1976, p.4). Chinese

Internet homophones have certain phonetic and morphological features (Chapter 3), and are

related to the lexical meaning of their base words. This chapter will examine the cohesion

function of Chinese Internet homophones based on the Leiden Weibo Corpus (LWC). We can

see that the lexical units are connected through the use of Chinese Internet homophones as

cohesive devices. The analysis in this chapter will also show the semantic relation between

different units within the Chinese Internet discourse.

Chapter 2 summarized the development of cohesion theory in the English and Chinese literature.

In this study, I use the framework of cohesion proposed by Hu et al. (2005) which is developed

from Halliday & Hasan (1976). The reasons for adopting this classification have been discussed

in Chapter 2. This classification includes grammatical cohesion and lexical cohesion.

Grammatical cohesion includes reference, ellipsis, substitution, and conjunction. Lexical

cohesion is comprised of repetition, synonymy/antonymy, hyponymy/meronymy, and

collocation. Table 4.1 provides a detailed summary of Hu et al. (2005)’s classification of

cohesion. It is used in the data analysis in Section 4.1 and Section 4.2.

59  

 

Table 4.1 Hu et al. (2005)’s Classification of Cohesion

Category Subcategory

Grammatical Cohesion

Reference

Personals Demonstratives Comparatives Endophoric:Anaphoric, cataphoric Exophoric

Ellipsis Nominal ellipsis Verbal ellipsis Clausal ellipsis

Substitution Nominal substitution Verbal substitution Clausal substitution

Conjunction Transitional words     Lexical Cohesion

Repetition Synonymy/Antonymy

Hyponymy/ Meronymy Collocation

In this chapter, the analysis will focus on homophones that serve as lexical and grammatical

cohesive devices. In what follows, Section 4.1 will examine the grammatical cohesion function

of Chinese Internet homophones, and Section 4.2 will investigate the lexical cohesion function of

Chinese Internet homophones.

4. 1 Grammatical Cohesion

According to Halliday & Hassan (1976) and Hu et al. (2005), grammatical cohesion includes

reference, ellipsis, substitution, and conjunction. The following section will explore the different

types of grammatical cohesion function of homophones in Chinese Internet discourse. Following

60  

 

the definition of each type of cohesion, examples from the data Leiden Weibo Corpus will be

provided.

4.2.1 Reference

Reference means that a participant or circumstantial element serves as another element’s referent.

If an element serves as a reference, it makes the information signaled for retrieval. When the

referential meaning is retrieved, “the cohesion lies in the continuity of reference, whereby the

same thing enters into the discourse a second time” (Halliday & Hasan, 1976, p. 31). Reference

includes personal reference, demonstrative reference, and comparative reference. Usually,

personal reference words refer to personal pronouns, such as you and him. Comparative

references refer to words with “comparative” meanings, such as same as and different.

Demonstrative references include definite articles (e.g. that) along with place and time

demonstrative words (e.g. there and now).  In Chinese, they are usually proximals, such as 这

(zhe4 ‘this’), 此 (ci3 ‘this’), and 这里 (zhe4li3 ‘here’) and distal-deixis, such as 那 (na4 ‘that’),

彼 (bi3 ‘that’), and 那里 (na4li3 ‘there’).

Reference can be divided into exophora and endophora depending on the relationship between

the referent, context, and discourse. The referent of an exophoric reference is found in the

context of situation. The referent of an endophoric reference lies in the surrounding text.

Furthermore, endophoric reference can be subcategorized into anaphora and cataphora, based on

the position of the referent within discourse. The referent of an anaphoric reference is in the

preceding text, while that of a cataphoric reference is in the following text.

61  

 

The following examples are drawn from the data for analysis. Gloss translation is provided to the

Chinese examples.33 Each example contains a unique message ID which can be found in the

Leiden Weibo Corpus.

The following example is about a leading man in a drama, which involves both demonstrative

and anaphoric reference.

[4: 1] (LWC, Message ID 3401883045887457)

1. 难怪 个个 都 喜欢上 李 大仁，

nan2gui4 ge4ge4 dou1 xi3huan1shang4 Li3 Da4ren3,

no doubt everyone all like Li Daren,

‘There is no doubt that everyone likes Li Daren.’

2. 介个 火星 好 男人， 也 只有 电视剧 会 出现 ，

jie4ge4 huo3xing1 hao3 nan2ren2, ye3 zhi2you3 dian4shi4ju4 hui2 chu1xian4,

this Mars good man, yet only television will appear,

‘ Yet this good Martian guy only appears in television,’

3. 因为 本 故事 纯粹 虚构. ...

yin1wei4 ben3 gu4shi4 chun2cui3 xu1gou4

because this story indeed fabricate

‘as this story is indeed fabricated…’

                                                                                                                         33 For abbreviations used in the glosses of the Chinese examples, please see Appendix.

62  

 

In line 2, 介个 (jie4ge4, 介 ‘interpose’, 个 ‘CL’, 介个‘this’), a homophone of 这个 (zhe4ge4

‘this /these’), derived from the Tianjin vernacular, a variation of the Northern Mandarin dialect.

The homophone 介个 refers to 李大仁 (Li3 Da4ren3, ‘a leading man in a Taiwan idol drama’).

介个 serves as a demonstrative reference and also an endophoric reference. The referent of 介个

is 李大仁. As the referent is in the preceding text, 介个 can be further categorized as an

anaphoric reference. The reference of the homophone 介个 creates cohesion, as it links the two

clauses into a cohesive unity.

Example [4:2] is a complaint from a bank clerk resulting from many people coming to his branch

before the Chinese New Year. It exemplifies exophoric, demonstrative, and personal reference.

[4:2] (LWC, Message ID 3402444914763617)

1. 人山 人海！！！ 人山 人海！！！ 人山 人海 啊！！！

ren2shan1 ren2hai3! ren2shan1 ren2hai3! ren2shan1 ren2hai3 a1!

people mountain people sea! people mountain people sea! people mountain people sea PRT!

‘A sea of people! A sea of people! A sea of people!’

2.介个 不 会 按 密码， 类个 不 会 填 单子，

jie4ge4 bu2 hui4 an4 mi4ma3, lei2ge4 bu2 hui4 tian2 dan1zi,

this NEG can press password, that NEG can fill form,

‘This [person] doesn’t know how to key in [his] password, that [person] doesn’t know how to

fill out the form,’

63  

 

3. 还有 一批 抱怨 排队 太久 的 人，

hai2you3 yi4pi1 bao4yuan4 pai2dui4 tai4jiu3 de1 ren2,

and a CL complain line up too long ASSC people,

‘and there is a group of people complaining that they are waiting in line too long.’

4.乃们 把 我 劈 成 几半 好么，

nai3men2 ba3 wo3 pi1cheng2 ji3ban4 hao3 me,

2nd PL BA 1stSG chop into several half OK,

‘Would you please chop me into several pieces?’

5. 谁 规定 过年 前 灰得34 来 银行 凑 一脚 哒？

shui2 gui1ding4 guo4nian2 qian2 hui1de3 lai2 yin2hang2 cou4 yi1jiao3 da1?

who set new year before have to come bank add a feet PRT?

‘Who told [you] that [you] must come to the bank before the [Chinese] New Year?’

6. 抱头 痛哭 啊 我

bao4 tou2 tong4 ku1 a1 wo3.

hug head cry hard PRT 1stSG.

‘I [am] hugging [my] head and crying sadly.’

In Line 2, 介个 (jie4ge4, 介 ‘interpose’, 个 ‘CL’, 介个‘this’) is a homophone of 这个 (zhe4ge4

‘this/these’) and 类个(lei4ge4, 类 ‘category’, 个 ‘CL’, 类个‘that’) is a homophone of 那个

(na4ge4 ‘that/those’). In Lin 4, 乃们 (nai3men2, 乃 ‘you’ in ancient Chinese, 们 ‘suffix for

plural personal pronoun’, 乃们 ‘you’) is a homophone of 你们 (ni3men2 ‘you’). 介个, 类个,

                                                                                                                         34灰得(hui1de3, 灰 ‘grey’, 得 ‘have to’ , 灰得 here it means ‘have to’) is a homophone of 非得 (fei1de3 ‘have to’).

64  

 

and 乃们 all serve as reference in the discourse. From the message, no referent can be found in

the discourse itself. By reading the message, this micro-blog user works at a bank and he was

complaining about his customers who came to the bank the day before the Chinese New Year.

Retrieving information from the context, 介个, 类个, and 乃们 all refer to this micro-blog user’s

customers. As the referent is in the context of situation, they all serve as exophoric reference.介

个 and 类个 are also demonstrative reference, and 乃们 is a personal reference. The use of 介个,

类个, and 乃们 specify their referents from the context and connect the text to form a unified

message.

The following example talks about “that thing” which is helpful at work, but there is no context

showing the referential meaning. This message displays exophoric and demonstrative reference.

[4:3] (LWC, Message ID 3402333038113555)

1.如果 我 上班 有 嘞个 东东 的话·· 不 晓得 好 给力 哟··

ru2guo3 wo3 shang4ban1 you3 lei4ge4 dong1dong de1hua4, bu4 xiao3de2 hao3 gei2li4 yo1.

if 1stSG work have that thing if, NEG know very great PRT.

‘If I had that thing at work, I can’t tell [you] how great it would be.’

2.可以 保暖 还 能 防 噪音· 一箭双雕 哈·

ke2yi2 bao3nuan3 hai2 neng2 fang2 zao4yin1. Yi2jian4shuang1diao1 ha1.

can keep warm and can prevent noise. One arrow two birds PRT.

‘[It] could keep me warm and prevent noise. [It would] kill two birds with one stone.’

65  

 

嘞个(le1ge4, 嘞‘PRT’, 个 ‘CL’, 嘞个 ‘that’) is a homophone of 那个(na4ge4 ‘that’). The

referential meaning of 嘞个 has to be found outside of the text.35 嘞个 servers as exophoric and

demonstrative reference. In this case, by reading one piece of the message, it is unlikely to

retrieve the specific information of what 嘞个 refers to. More information is needed from the

context of situation.

In summary, reference includes personal reference, demonstrative reference, and comparative

reference. It can also be classified into exophora and endophora. In an exophoric reference, the

referent is in the context of situation. In an endophoric reference, the referent can be either found

in the preceding text (anaphora) or the following text (cataphora). In the data, the majority of the

Chinese Internet homophones display personal, demonstrative, anaphoric, and exophoric

references. Chinese Internet homophones can serve as cohesive devices, linking the referential

meaning of the reference to the referent, and create cohesion within discourse.

4.2.2 Ellipsis

Ellipsis is “something left unsaid”, but “understood nevertheless”(Halliday & Hasan, 1976,

p.142). “Where there is ellipsis, there is a presupposition, in the structure, that something is to be

supplied, or ‘understood’ ” (Halliday & Hasan, 1976, p.144). This phenomenon is very common

in Chinese. Ellipsis is characterized as having syntactic recoverability, that there is a

corresponding completed item which presents the unsaid information. The elliptical item and the

corresponding item are semantically the same. The elliptical item leaves a specific structural slot                                                                                                                          35 For this message, there is no clear context showing the exact referent.

66  

 

to be filled by the corresponding item (Halliday & Hasan, 1976, p. 143). Moreover, this syntactic

recoverability and semantic consistency can be achieved through a certain pragmatic

environment. Generally, ellipsis in Chinese is more widely used than that in English (Zuo, 1995,

p.41).

Ellipsis, by omitting some components in discourse, reduces redundancy and still represents the

semantic information of the elliptical items. It is a grammatical method to make discourse

compact and cohesive. There are three types of ellipsis, namely nominal ellipsis, verbal ellipsis,

and clausal ellipsis. The elliptical item is consistent with the presupposition, and they are in the

same grammatical class. Usually, the elliptical item is easy to recover from the previous text. In

Chinese, the subject is commonly omitted (Zuo, 1995, p.41).

The following three examples are all from the LWC. Example [4:4], shown below, shows a

micro-blog user’s feelings about their travel in Australia ending soon. It presents nominal ellipsis,

or specifically subject ellipsis.

[4:4]36 (LWC, Message ID 3399871598226546)

                                                                                                                         36 This is an excerpt of the original message.

67  

 

1. 偶 的 墨尔本 之旅 很 快 就 结束 了 哇。

ou3 de1 Mo3er3ben3 zhi1lv3 hen3kuai4 jiu4 jie2su4 le1 wa1.

1stSG POSS Melbourne ASSO trip very quick just end CRS PRT.

‘My trip to Melbourne will end soon.’

2. 舍 不 得 舍 不 得。

________she3 bu4 de2 she3 bu4 de2.

abandon NEG CSC abandon NEG CSC.

‘[ I ] am very reluctant to go.’

偶 (ou3 ‘even’, it means ‘I’ here) is the homophone of 我 (wo3 ‘I’). As the subject, 偶 is omitted

in the second line. The unsaid subject in Line 2 can be recovered from “偶” in the preceding text.

The homophone 偶 serves as the source to supply the missing information.

In example [4:5], a blogger “cold jest” talks about a little girl who was eating noodles. It contains

nominal (subject) ellipsis.

[4:5] (LWC, Message ID 3399769395789797)

68  

 

1. #冷笑话# 对面 的 小 MM 吃 面 真 浪费。

leng3 xiao4hua4: dui3mian4 de1 xiao3 mei3mei2 chi1 mian4 zhen2 lang4 fei4.

cold jest: opposite ASSC little girl eat noodle very wasteful.

‘A cold jest: the little girl sitting opposite to me really wasted her noodles.’

2. _____面条 扒拉 两口 就 走人。

____mian4tiao2 ba1la1 liang3kou3 jiu4 zou3 ren2.

noodles stir two CL then left.

‘[ The little girl ] only had a few bites and then left.’

3.于是 我 正义 的 把 那 碗 面 倒给 路边 看 起来 很 饿 的 野猫。

yu2shi4 wo3 zheng4yi4 de1 ba3 na4 wan3 mian4 dao4gei3 lu4bian1 kan3qi3lai3 hen3e4 de1 ye3mao1.

so 1stSG righteous ADV BA that CL noodle pour street look like very hungry ASSC stray cat.

‘So righteously, I poured the bowl of noodles on the side of the road for the hungry-looking stray

cat.

4. 一会儿， 小 MM 回来 了, 手里 拿着 一 瓶 水。

yi2 hui4r, xiao3 mei3mei2 hui2lai2 le1, shou3li3 na2zhe yi4 ping2 shui3.

soon, little girl come back CRS, hand hold a CL water.

‘After a little while, the little girl came back with a bottle of water in her hand.’

5. ______一 脸 狐疑 的 看着 那 个 空碗…

_____ yi4 lian3 hu2yi2 de1 kan4zhe na4ge4 kong1wan3.

a face suspicious ADV look that CL empty bowl.

‘[The little girl] looked at the empty bowl with a suspicious face.’

6.那 一刻， 我 只想 当 个 埋头 吃面 的 路人甲…

69  

 

na4 yi2ke4, wo3 zhi3xiang3 dang1 ge4 mai2tou2 chi1mian4 de1 lu4ren2jia3.

that moment, 1stSG only want be CL face down eat noodle ASSC passer-by.

‘At that moment, I just wanted to be [a] passer-by eating noodles with my face down.’

MM (mei3mei2 ‘girl’) is a homophone of 美眉 (mei3mei2 ‘girl/beautiful girl’). 小 MM (xiao3

mei3mei2 ‘little girl’) means ‘a little girl’ here. As the two empty structural slots in Line 2 and

Line 5 shown above, there are two places where 小 MM is omitted. Subject ellipsis is also a type

of nominal ellipsis. Subject ellipsis is very commonly used in Chinese to make discourse

compact and cohesive.

In example [4:6] the mirco-blog user talks about writing micro-blogs to his friend “Le-Shu-Yan”.

It presents nominal ellipsis after numbers and a classifier (CL) tiao.

[4:6] (LWC, Message ID 3399682279759812)

1. 对@Le-淑-Yan 说:

dui4 Le4 Shu1yan shuo1:

to Le Shuyan speak:

‘[I] speak to Le Shuyan:’

70  

 

2. 第 99 条 围脖 给 你,

di4 99 tiao2 wei2bo2 gei3 ni3,

99th CL Weibo for 2nd SG,

‘the 99th Weibo is for you.’

3. 以后 第 199. 299. 399. 499. 599. 699. 799. 899. 999 条______都 给你

yi3hou4 di4 199 299 399 499 599 699 799 899 999 tiao_____dou1 gei3 ni3.

in the future 199th, 299 th, 399 th, 499 th, 599 th, 699 th, 799 th, 899 th, 999th, CL all for 2nd SG.

‘In the future the 199 th, 299 th, 399 th, 499 th, 599 th, 699 th, 799 th, 899 th, 999th [Weibo] are all for

you’

This micro-blog user posted a Weibo (micro-blog) message to his friend Le Shuyan to show his

affection toward her.37 围脖 (wei2bo2 ‘scarf’, here means ‘micro-blog’) is a homophone of 微博

(wei1bo2 ‘micro-blog’). In Line 2, 第 99 条围脖 (di4 jiu3shi2jiu4 tiao2 wei2bo2 ‘the 99th CL

Weibo message’) appears in the pattern of “number + CL+noun”. In Line 3, the clause 第 199,

299, 399, 499, 599, 699, 799, 899, 999 条(di4 yi4bai3jiu3shi2jiu3, er4bai3jiu3shi2jiu3,

san1bai3jiu3shi2jiu3, si4bai3jiu3shi2jiu3, wu3bai3jiu3shi2jiu3, liu4bai3jiu3shi2jiu3,

qi1bai3jiu3shi2jiu3, ba1bai3jiu3shi2jiu3, jiu3bai3jiu3shi2jiu3 tiao2 ‘the 199th, 299th, 399th, 499th,

599th, 699th, 799th, 899th, 999th CL’) follows the same pattern. As the noun 围脖 has been

mentioned in Line 2 ‘the 99th CL Weibo message’, it is omitted at Line 3, but the “unsaid”

information is easy to be retrieved.

                                                                                                                         37 The numeric letter 9 in Chinese has the meaning of ‘forever’. From the words 99, 199, 299…in this message, it reveals that this micro-blog user wishes the relationship between him and his friend Le Shuyan will last forever.

71  

 

Ellipsis includes nominal ellipsis, verbal ellipsis, and clausal ellipsis. In this section, I focused

on nominal ellipsis, as it is the only ellipsis type found in my data. I have also found that most of

the nominal elliptical items are subjects, which is consistent with Zuo (1995)’s conclusion that

subjects are more commonly omitted in Chinese. These examples show that the elliptical items

can be retrieved from the preceding homophones. Homophones serve as the source of elliptical

items and contribute to cohesion within discourse.

 

4.2.3 Substitution

Substitution means to use one item to replace another. The meaning of the substitution item is

drawn from the item that has been replaced. Substitution includes nominal substitution, verbal

substitution, and clausal substitution. Hu et al. (2005) argued that it is controversial whether

there are items in Chinese that are similar to those in English, which can substitute nouns, verbs,

or clauses. For nominal substitution in English, if the same noun appears again, it will be

substituted by a related syntactic item, such as one. For instance, in the example “This shirt is

too dark. Would you please find me a yellow one?”, one substitutes shirt. Usually, in Chinese,

的 (de1 ‘nominalizer’) can be used as a nominalizer. In the pattern “noun/personal

pronoun/adjective + (的)+ noun”, 的 can be used to substitute the noun after 的. “Noun/personal

pronoun/adjective +的” is equivalent to a noun phrase in the “de (的) structure”.

To give an example of the “de (的) structure”:

72  

 

[4:7]

这 件 衬衫 颜色 太 深 了。能 给 我 找 一件 黄 的 吗？

zhe4 jian4 chen4shan1 yan2se4 tai4 shen1 le1. neng2 gei3 wo3 zhao3 yi2 jian4 huang2 de ma?

this CL shirt color too deep CRS. can for 1stSG find one CL yellow NOM Q?

‘The color of this shirt is too dark. Would you find me a yellow one?’

衬衫 (chen4shan1 ‘shirt’) is mentioned in the first sentence. In the second one, after the

adjective word 黄 (huang2 ‘yellow’), 的 substitutes 衬衫 from the proceeding text. 的 acts as a

cohesive device through nominal substitution.

Substitution is not very common in the data and the only type of usage of homophones as

substitution is 滴. As in Examples [4:8], the homophone 滴 serves as a nominal substitution

device.

[4:8] (LWC, Message ID 3399828346705296)

1. Chanel 最新 紫色，银联 的 timeless classic flap 喔， 喜欢 的 联系 吧，

Chanel zui4xin2 zi3se4, yin2lian2 de1 timeless classfic flap wo1, xi2huan1 de1 lian2xi4 ba1;

Chanel newest purple, Union Pay ASSC timeless classic flap PRT, like NOM contact PRT;

‘A Chanel [bag], in the latest purple color [comes with a] timeless classic flap from Union Pay;

Please contact [us] if you like it;’

2. 有 喜欢 银色 滴___ 也 可以 联系 喔

you2 xi3huan1 yin2se4 di1____ye3 ke3 yi3 lian2xi4 wo1.

73  

 

has like silver NOM also can contact PRT.

“those who like silver ones can also contact [us].”

滴(di1 ‘drop’, here it is used as a nominalizer) is a homophone of 的(de1 ‘nominalizer’). After

银色 (yin2se4 ‘silver’), 滴 replaces Chanel (bag). The homophone 滴 serves as a cohesive

device, linking the two clauses.

In English, if the same verb is used again, but replaced with a related syntactic item, such as do

in the text, it is verbal substitution. In Chinese, certain items, such as 是(shi4 ‘so’) and 做 (zuo4

‘do’), can substitute verbs. However, there is no such homophone found in the data. Clausal

substitution refers to the phenomenon that a repeating clause is substituted by a related syntactic

item, such as so. In Chinese, there is no equivalent usage of so as in English. Only in certain

cases, items such as 是(shi4 ‘so’) can substitute a clause. As there is no such usage of verbal

substitution or clausal substitution of Chinese Internet homophones found in the data, no specific

examples will be given in this thesis.

To summarize, substitution includes nominal substitution, verbal substitution, and clausal

substitution. Nominal substitution by using 的 as a nominalizer is commonly found in Chinese.

The homophone 滴 can occur as a substitution item making clauses linked to each other.

74  

 

4.2.4 Conjunction

“Conjunctive elements are cohesive not in themselves but indirectly by virtue of their specific

meanings,” and “they express certain meanings which presuppose the presence of other

components in the discourse” (Halliday & Hasan, 1976, p.226). Usually, conjunctive items are

transitional words, indicating logical relations between different clauses. These logical relations

include temporal, causal, conditional, and so on. Conjunctive items include conjunctions,

correlative adverbs, and other types of words playing the role of conjunction. In discourse,

sometimes there is no specific item that indicates a specific logical relation. This type of

conjunction is called an implicit conjunction, or parataxis conjunction. All the examples of

conjunction in this section are explicit.

An example [4:9] from Hu et al. (2005) showing conjunction that indicates a progressive relation

is provided below.

[4: 9] (Hu et al., 2005, p.194)

1. “……我 从没 见过 他 办过 一 件 事 要 花 半天 工夫！

…wo3 cong2 mei2 jian4guo4 ta1 ban3guo4 yi2 jian4 shi4 yao4 hua1 ban4tian1 gong1fu!

…1stSG even NEG seen 3rdSG deal one CL thing would take half day effort!

‘… I have never seen him take half a day to get anything done,

2. 何况 是 那么 一点 小事，

he2kuang4 shi4 na4me yi4dian3 xiao3shi4,

75  

 

moreover be that a bit little thing,

‘[and] moreover, for such a minor issue.’

3. 他 只要 眉头 一 皱， 办法 就 全 有 了！……”

ta1 zhi3yao4 mei2tou2 yi2 zhou4, ban4fa3 jiu4 quan2 you3 le! …

3rdSG only eyebrow frown, solution then all have CRS! …

‘Only if he frowns, solutions will arrive! ...’

The example tells that the man can solve problems very quickly. 何况 (he2kuang4 ‘moreover’)

connects 那么一点小事 (na3me yi4dian3 xiao3shi4 ‘such a minor issue’) and add the

information that he can come up with a solution in a second. The conjunctive item 何况

expresses a progressive relation.

The following example is drawn from the LWC. There were not many examples found of this

type. There is only one homophone 可素 showing a transitive relation in [4:10].

[4:10] (LWC, Message ID 3402332436848113)

1. 腊月 二十三 啊！ 小年 啊！

la4yue4 er4shi2san1 a1! xiao3nian2 a1!

lunar year’s December 23rd PRT! Spring Festival Eve PRT!

‘December 23rd [is] the Chinese Lunar New Year! Spring Festival Eve!’

76  

 

2. 可素 我们 不 放假！！！ 介38 都 是 神马39 世道 啊！！！

ke2su4 wo3men bu2 fang4jia4! jie4 dou1 shi4 shen2ma3 shi4dao4 a1!

but 1stPL NEG have holiday! this all be what world PRT!

‘But we don’t have a holiday! What kind of a world is this!’

可素(ke3su4, 可‘but’, 素 ‘plain’, 可素 means ‘but’ here) is a homophone of 可是(ke3shi4 ‘but’).

It links the message with an explicit transition. 可素 serves as a cohesive device connecting its

preceding and following texts.

In this section, through the temporal, causal, conditional and other relations, Chinese Internet

homophones connect elements in a micro-blogging message and serve as cohesive devices. But

the usage of Chinese Internet homophones as conjunctions is not very common compared to

other types of grammatical cohesion in the data.

4.3 Lexical Cohesion

Lexical cohesion is the effect achieved through the choice of lexical items and operates within

the lexis (Halliday & Matthiessen, 2004, p. 535). The following sections will analyze lexical

cohesion of Chinese Internet homophones, namely from repetition, synonymy/antonymy,

hyponymy/meronymy, and collocation.

                                                                                                                         38介(jie4 ‘interpose’), the homophone of 这 zhe4 ‘this’, means ‘this’. 39 神马(shen2ma3, 神 ‘magic’, 马 ‘horse’), a homophone of 什么 shen3me1 ‘what’, means ‘what’.

77  

 

4.3.1 Repetition

Repetition refers to repeating a lexical item that has appeared in preceding text. Each repeated

item contains substantially the same ideational meaning40. Repetition is the simplest way to

achieve lexical cohesion. Repeated lexical items include simple repetition and inflectional

repetition (e.g., eat, eating, ate, and eaten; apple and apples). Chinese lacks morphological

changes and therefore inflectional repetition is very rare. These repeated words are usually the

keywords in a sentence or discourse.

Lexical repetition plays a very important role within discourse. It functions as a method of

cohesion and makes information exchange more efficiently. When it comes to communication,

there cannot always be new information, and this sometimes necessitates repetition. Additionally,

from a rhetorical consideration, repetition keeps the meaning consistent, and it will also attract

readers' attention. Therefore, readers will remember repeated content more easily and be more

strongly influenced by the writing (Liao, 2004, p. 117).

The following three examples are all from the LWC. They display cohesion by repetition of the

homophones.

Example [4:11] talks about a creepy man by repeating the homophone ws three times.

[4:11] (LWC, Message ID 3401488479999787)

                                                                                                                         40 For ideational meaning, please see Chapter 2.

78  

 

1. ws 男 看 ws 的 图片 竟然 要 ws 大学 男 解释，

wei3suo3 nan2 kan4 wei3suo3 de1 tu2pian4 jing4ran2 yao4 wei3suo3 da4xue2 nan2 jie3shi4,

creepy man look creepy ASSC picture unexpectedly want creepy university man explain,

‘The creepy man looked at the creepy picture and unexpectedly, he wanted the creepy

university man to explain it,’

2. 然后 ws 的 笑 了.

ran2 hou4 wei3suo3 de1 xiao4 le1.

then creepy ADV smile PFV.

‘and then [he] laughed creepily.’

3.可恨 的 是 旁边 还有 几 个 特 清纯 的 妹纸41

ke3hen4 de1 shi4 pang2bian1 hai2 you3 ji3 ge4 te4 qing1chun2 de1 mei4 zhi3.

hateful MON be besides and have several CL extremely innocent ASSC girls.

‘The terrible thing is that there were several extremely innocent girls beside [him].’

Ws (wei3suo3 ‘creepy’) is a homophone of 猥琐 (wei3suo3 ‘creepy’). The word ws appears four

times in this message and shows the man is very creepy. The cohesive device ws makes the

theme clear and the discourse cohesive.

Example [4:12] talks about the party and the government in China. It shows repetition of the

homophones d and ZF.

[4:12] (LWC, Message ID 3403229884397184)                                                                                                                          41妹纸(mei4 zhi3,妹‘sister’,纸‘paper’), a homophone of 妹子 mei4zi3 ‘sister’, means ‘sister’.

79  

 

1. 以前 一直 呼吁 希望 d 和 ZF 为 老百姓 做 点 什么，

yi3qian2 yi4zhi2 hu1yu4 xi1wang4 dang3 he2 zheng4fu3 wei4 lao3bai3xing4 zuo4 dian3 shen3me,

before always advocate hope party and government for civilians do a bit what,

‘Before, [I] always hoped and advocated that the party and the government would do

something [good] for the people.’

2. 现在 后悔 了，

xian4zai4 hou4hui3 le1.

now regret PFV.

‘Now [ I] have regretted [it].’

3. 只要 d 和 ZF 别 再 对 老百姓 做 什么 了

zhi3yao4 dang3 he3 zheng4fu3 bie2 zai4 dui4 lao3bai3xing4 zuo4 shen3me le,

only party and government NEG again toward civilians do what CRS,

“If only the party and the government wouldn’t do anything more to the people,’

4. 就 谢天谢地 了

jiu4 xie4tian1xie4di4 le1.

then thank goodness CRS.

‘I would be grateful.’

In this message, d (dang3 ‘party’) is a homophone of 党 (dang3 ‘party’) and ZF (zheng4fu3

‘government’) is a homophone of 政府(zheng4fu3 ‘government’). Because of Internet censorship

in China, the actual words of 党 and 政府 sometimes cannot be posted online. As a result,

80  

 

Internet users use d and ZF as replacement words. Through the repetition of d and ZF makes the

message lexically cohesive.

Example [4:13] is a New Year greeting message, showing repetition of the homophone 兔 in

several idioms and allusions, indicating the Year of the Rabbit.

[4:13] (LWC: Message ID 3404670880432789)

1. 姜 你 军42， 蒜 你 狠43， 我 兔 故 纳 新 算 得 准。

jiang2 ni3 jun1, suan4 ni3 hen3, wo2 tu4 gu4 na4 xin1 suan4 de2 zhun3.

ginger 2ndSG general , onion 2ndSG ruthless, 1stSG rabbit old take new calculate CSC accurate.

‘The prices of ginger and onions are killing me. “Out with the old, in with the new”, and I plan

everything well.’

2. 糖 高宗44， 豆 你 玩45，我 大 展 宏 兔 铁 臂 腕！

tang2 gao1zong1, dou4 ni3 wan2, wo3 da4 zhan3 hong2 tu4 tie2 bi4 wan3!

sugar Gaozong, bean 2ndSG play, 1stSG big spread big rabbit steel arm wrist!

‘The prices of sugar and beans are crazy; I [can] succeed [if] I am strong!’

3. 谁 说 肉 的 理想 只 有 白菜 的 命？

shui2 shuo1 rou4 de1 li3xiang2 zhi3 you2 bai2cai4 de1 ming4?

who say meat ASSC dream only have cabbage ASSC destiny?                                                                                                                          42姜你军 (jiang1 ni3 jun1 ‘ginger your general’), a homophonic phrase of 将你军(jiang1 ni3 jun1 ‘to kill your General (a term from the Chinese chase game)’), means that the price of ginger is killing people. 43蒜你狠 (suan4 ni3 hen3 ‘union you ruthless’), a homophonic phrase of 算你狠 (suan4 ni3 hen3 ‘you are malicious’), means that the price of unions is killing people. 44糖高宗(tang2 gao1zong1 ‘sugar Gaozong’), a homophonic phrase of 唐高宗 (tang2 gao1zong1 ‘Emperor Gaozong of the Tang Dynasty’), means that the price of sugar is very high. 45豆你玩(dou4 ni3 wan2 ‘bean you play’), a homophonic phrase of 逗你玩(dou4 ni3 wan2 ‘it is a joke’), means that the price of beans is crazy.

81  

 

‘Who says dreams can’t come true?’

4. 只要 “兔兔” 是 道46， 就 能 扬 眉 “兔” 气，

zhi3yao4 “tu4 tu4” shi4 dao4, jiu4 neng2 yang2 mei2 tu4 qi4.

only if “rabbit rabbit” be Dao, then can raise eyebrow rabbit air.

‘Only by finding the Dao everywhere, can you become proud and elated.’

5. 恭祝： 龙 年 吉 祥，

gong1zhu4: long2 nian2 ji2 xiang2.

sincerely wish: dragon year lucky.

‘[I] sincerely wish [you] a good luck in the Year of the Dragon.’

6. 不 求 腾 云 驾雾， 只 求 赛过 金龙！

bu4 qiu2 teng2 yun2 jia4 wu4, zhi3 qiu2 sai4 guo4 jin1long2!

NEG hope mount cloud ride mist, only hope surpass golden dragon!

‘Do not try to mount the clouds and ride the mist [like a god], but I hope that you can surpass the

golden dragon [and become wealthy]!’

This New Year greeting message uses 兔 (tu4 ‘rabbit’) as a homophone to replace the words in

the idioms. 兔故纳新(tu4 gu4 na4 xin1, 兔 ‘rabbit’, 故 ‘old’, 纳 ‘take’, 新 ‘new’) a homophone

of 吐故纳新(tu3 gu4 na4 xin1 吐 ‘exhale’, 故 ‘old’, 纳 ‘take in’, 新 ‘new’, 吐故纳新 ‘out with

the old, in with the new’), means ‘out with the old, in with the new’. 大展宏兔 (da4 zhan3

hong2 tu4, 大‘big’, 展 ‘spread’, 宏 ‘big’, 兔‘rabbit’), a homophone of 大展宏图 (da4 zhan3

hong2 tu2, 大 ‘big’, 展 ‘spread’, 宏 ‘big’, 图 ‘picture’, 大展宏图‘to succeed’ ), means ‘to

                                                                                                                         46道 (dao4 ‘Dao’) here refers to 道场 (dao4chang3 ‘Bodhimanda’), a term from Buddhism, meaning the “position of awakening”.

82  

 

succeed’. 兔兔是道(tu4 tu4 shi4 dao4, 兔 ‘rabbit’,兔 ‘rabbit’, 是 ‘be’, 道 ‘Dao’), a homophone

of 处处是道(chu4 chu4 shi4 dao4, 处处 ‘everywhere’, 是 ‘is’, 道 ‘Dao’, 处处是道 ‘Dao is

everywhere’), means ‘Dao is everywhere’. 扬眉兔气(yang2 mei2 tu4 qi4, 扬 ‘raise’, 眉

‘eyebrow’, 兔 ‘rabbit’, 气 ‘air’ ) a homophone of 扬眉吐气(yang2 mei4 tu3 qi4, 扬 ‘raise’, 眉

‘eyebrow’, 吐 ‘exhale’, 气 ‘air’, 扬眉吐气 ‘to feel proud and elated’) means ‘ to feel proud and

elated’. Through repetition of 兔 four times，this message is cohesively linked. Thus the

message is clear and prominent, looking back upon the Year of the Rabbit and then making

wishes for the Year of the Dragon (the coming New Year). The lexical repetition is used for

cohesion in discourse.

4.3.2 Synonymy/Antonymy

Synonymy and antonymy also play an important role in cohesion. Synonyms are

morphologically different, but with the same or a close meaning. In discourse, synonyms can be

repeated in a unit of semantic meanings and supplement the preceding content. Synonymy makes

different components semantically linked to each other. In addition, from a rhetorical point of

view, the use of synonyms can avoid repetition and prevent a monotonous style in the discourse.

It makes the expression varied and vivid. Antonymy is established between two lexical or more

items with opposite meanings. The use of antonyms generates a contrastive effect, links

discourse, and achieves cohesion.

83  

 

The following examples are from the LWC, showing cohesion created by synonymy or

antonymy.

Example [4:14] talks about the Chinese Communist Party’s policy towards Hong Kong. It shows

cohesion created by synonymy.

[4:14] (LWC, Message ID 3404225311063790)47

1. 对于 香港 这个 地方 的 优越感，

dui4 yu2 xiang1gang3 zhe4ge4 di4fang2 de1 you1yue4gan3,

as to Hong Kong this CL place ASSC superiority feeling,

‘Regarding Hong Kong, this place’s sense of superiority’

2. 完全 是 中共 自己 惯 出来 的。

wan2quan2 shi4 Zhong1gong4 zi4ji3 guan4 chu1lai2 de1.

total be Chinese communist party itself spoil out PRT.

‘has been totally spoiled by the Chinese Communist Party.’

3. 78 年 以前 是 只有 这个 对外 的 窗口，

78 nian2 yi3qian2 shi4 zhi2you3 zhe4g4 dui4wai4 de1 chuang1kou3,

78 year ago be only have this CL external ASSC window,

‘Before 1978, it was the only window out.’

4.改革 后 香港 的 作用 和 地位 肯定 要 下降，

gai3ge2 hou4 Xiang1gang3 de1 zuo4yong4 he2 di4wei4 ken3ding4 yao4 xia4 jiang4,

reform later Hong Kong ACCS role and status definite would decrease,                                                                                                                          47 This is an excerpt of the original message.

84  

 

‘after the reform, Hong Kong’s role and status will definitely decrease.’

5.但 TG 认为

dan4 tian1gong4/tu3gong4 ren4wei2

but Celestial Kingdom of the Communist Party/ Communist Bandits think

‘But the Celestial Kingdom of the Communist Party/ Communist Bandits thinks that’

6. 要是 在 自己 手里 下降 会 显得 没 面子，

yao4shi4 zai4 zi4ji3 shou3li3 xia4jiang4 hui4 xian3de2 mei2 mian4zi,

if at itself hand decrease would seem NEG face,

‘if a recession happens under their supervision, they will lose face.’

7. 所以 资源 尽量 倾斜 港澳， 造就 出 这么 一 个 极端 地方。

suo3yi3 zi1yuan3 jin4liang4 qing1xie3 Gang3ao4, zao4jiu4 chu1zhe4me yi2ge4 ji2duan1 di4fang1.

so resource try to lean Hong kong & Macao, build out such a CL extreme place.

‘So they tried their best to invest lots of resources in Hong Kong and Macao to build such a

special place.’

中共(zhong1gong4) refers to ‘the Chinese communist party’. TG (tian1gong4/tu3gong4) is a

homophone of 天共 (tian1gong4 ‘Celestial Kingdom of the Communist Party’) or 土共

(tu3gong4 ‘Communist Bandits’). The micro-blog user used 中共 and TG as a pair of synonyms,

referring to the Chinese Communist Party. 中共 and TG function as synonymy to connect the

message.

85  

 

Example [4:15] uses “tragedy” and “comedy” to talk about life, showing cohesion created by

antonymy.

[4:15] (LWC, Message ID 3400563930754870)48

1. 生活 是 自己 的 杯具, 别人 眼里 的 洗具。

sheng1huo2 shi4 zi4ji3 de1 bei1ju4 bie2ren2 yan3li3 de1 xi3ju4.

life be your own ASSC tragedy, other people eye ASSC comedy.

“Life is a tragedy to you, but a comedy to others.”

In the message,杯具 (bei1 ju4, 杯‘cup’, 具 ‘tool’, 杯具 ‘cup’, here it refers to ‘tragedy’), a

homophone of 悲剧 (bei1ju4 ‘tragedy’), means ‘tragedy’. 洗具 (xi3ju4, 洗 ‘clean’, 具 ‘tool’, 洗

具 ‘cleaning set’, here it refers to ‘comedy’), a homophone of 喜剧 (xi3ju4 ‘comedy’), means

‘comedy’. The two homophones 杯具 and 洗具 are a pair of antonyms. The antonymy created

by 杯具 and 洗具 makes the text cohesive.

In example [4:16], the micro-blog user talks about finding beautiful girls on the Internet. It

displays cohesion through synonymy and antonymy.

[4:16] (LWC, Message ID 3402490597102154)

1. 美眉 几 时 有， 上机 问 网友，

mei3mei2 ji3 shi2 you3, shang4ji1 wen4 wang3you3,

                                                                                                                         48 This is an excerpt of the original message.

86  

 

beautiful eyebrow when have, use computer ask net-friend,

‘When there will be beautiful girls? I am using a computer to ask friends online.’

2. 不知 网上 佳丽， 此 妹 是 何人？

bu4 zhi1 wang3shang4 jia1li4, ci3 mei4 shi4 he2ren2?

NEG know Internet beautiful girls, this girl be who?

‘I don’t know [any] beautiful girls online. Who is this girl?’

3. 我 欲 下线 归 去，惟 恐 天 南 海 北， 远处 不 堪 寻，

wo3 yu4 xia4xian4 gui1qu4, wei2 kong3 tian1nan2 hai3 bei3, yuan3chu4 bu4kan1 xun2.

1stSG want off line go back, only afraid sky south sea north, far NEG can search.

‘I want to be offline and go home, but [I am] afraid that I am not able to search very far or all

over the country.’

4. 查找 选 芳名， 何似 在 网吧.

cha2zhao3 xuan3 fang1ming2, he4si4 zai4 wang3ba1.

search and find select girls’ name, NEG compare to at Internet bar.

‘In terms of finding girls, it can’t compare to [being] in an Internet bar.’

5. 打 电话， 记 传呼， 单 照 付， 不 应 心 疼，

da3 dian4hua4, ji4 chuan2hu1, dan1 zhao4fu4, bu4 ying1 xin1teng2.

make phone call, write down pager, bill still pay, NEG should care.

‘[I] make phone calls, write down the pager messages, pay bills, [and] shouldn’t care [about

money].’

5. 何时 不 想 聚 时 缘？

he2shi2 bu4 xiang3 ju4 shi2 yuan2 ?

87  

 

when NEG think get together time destiny?

‘When [do you] not think about destiny when [you think about] getting together?’

6. 月 有 阴 晴 圆 缺， 网 有 恐龙 青蛙49，

yue4 you3 yin1 qing2 yuan2 que2, wang3 you3 kong3 long2 qing1wa1.

moon has cloudy sunny round unround, Internet has dragon frogs.

‘ The moon can be bright or dim, full or waning, and there are ugly-looking girls and boys on

the Internet.’

7. 此 事 古 难 全。

ci3 shi4 gu3 nan2 quan2.

this thing ancient hardly perfect.

‘There has been nothing perfect since ancient times.’

8. 但愿 好 美眉 , 千 里50 共 聊 天。

dan4yuan4 hao3 mei3mei2, qian1 li3 gong4 liao2 tian1.

Wish good beautiful girl, thousand Li together chat.

‘[I] only wish to have a nice beautiful girl chatting with me, [even if we are] thousands of miles

away.’

美眉 (mei3mei2 美 ‘beautiful’, 眉 ‘eyebrows’, 美眉 ‘beautiful girl’) is a homophone of 妹妹

(mei4 mei4 ‘sister/girl’). The synonymy of 美眉, 佳丽(jia1li4 ‘beautiful girl’), and 妹 (mei4

‘sister/girl’) not only shows the user’s eagerness to find a beautiful girl on the Internet, but it also

avoid monotonous duplication. In addition, 美眉 and 恐龙 (kong3long2 ‘dragon’, it refers to

                                                                                                                         49青蛙 (qing1wa2 ‘frog’) is used to refer ugly-looking boys by Chinese Internet users. 50里(li3 ‘Li’) is a Chinese measure word for distance. 1 Li equals 500 Meters.

88  

 

‘ugly-looking girls’) are antonyms, showing a contrastive effect. The use of synonymy and

antonymy makes the message more cohesive.

4.3.3 Hyponymy/Meronymy

Hyponymy is the relation between specific lexical items and their superordinate. Meronyms and

their superordinate have a relation of “parts and whole”: A belongs to B, and A is a component

of B. For example, head, neck, and leg are meronyms of the superordinate body. The relation of

hyponyms and their superordinate is: the semantic field of A is included in that of B. Normally,

the meaning of a superordinate is general and abstract, whereas that of its hyponyms is specific.

For instance, apple, pear, and banana are hyponyms of the superordinate fruit. In discourse,

cohesion can be achieved by using either hyponyms/meronyms, or hyponyms/meronyms with

their superordinate.

The following example from Hu et al. (2005) shows meronymy in a Chinese novel. Cohesion is

achieved through different words of human body parts.

[4:17] (Hu et al., 2005, p. 194)

1. 她 的 脸 色 现在 又 飞 红了, 她 的 眼 光 迷乱，

ta1 de1 lian3 se4 xian4zai4 you4 fei1 hong2 le1, ta1 de1 yan3 guang1 mi2luan4,

3rdSG POSS face now again fly red CRS, 3rdSG POSS eye sight confuse,

‘Her face was blushing, her eyes were confused,’

2. 她的 胸部 很 剧烈 一起一伏。

89  

 

ta1de1 xiong1bu4 hen3 ju4 lie4 yi4qi3yi4fu2.

3rdSG POSS chest very extreme rise and fall.

‘and her chest rose and fell heavily.’

3. 突然 她 伸开 了 两臂。

tu1ran2 ta1 shen1kai1 le1 liang3 bi4.

all of a sudden 3rdSG open PFV two arms.

‘All of a sudden, she opened her arms.’

4. 雷 参谋 抢上 一 步，

Lei2 Can1mou2 qiang3shang4 yi 2bu4;

Lei Chief-of-staff take a step;

‘Chief-of-staff Lei took a step;’

5. 吴 少奶奶 便 像 醉 迷 似的 扑在 雷参谋 胸 前，

Wu2 Shao4nai3nai3 bian4 xiang4 zui4mi2 si4de pu1zai4 Lei2 Can1mou2 xiong1 qian2,

Wu lady then seem drunk like fall on at Lei Chief-of-staff chest front,

‘Lady Wu fell on the chest of Chief-of-staff Lei like a drunken person’

6. 她 的 脸 恰 靠 在 雷 参谋 肩头。

ta1 de1 lian3 qia4 kao4 zai4 Lei2 Can1mou2 jian1 tou2.

3rdSG POSS face just rest on Lei Chief-of-staff shoulder.

‘with her face just resting on Chief-of-staff Lei’s shoulder.’

7. 雷 参谋 俯下 头去， 两 个 嘴唇 接在 一起。

Lei2 Can2mou2 fu3xia4 tou2qu4, liang3ge4 zui3chun2 jie1zai4 yi4qi3.

Lei Chief-of-staff lower head, tow CL lips connect together.

90  

 

‘Chief-of-staff Lei lowered his head, and then the two [people’s] lips came together.’

身体(shen1ti3 ‘body/human body’)

脸 眼 两臂 胸部/胸 嘴唇 (lian3 ‘face’) (yan3 ‘eyes’) (liang3bi4 ‘arms’) (xiong1bu4/xiong1 ‘chest’) (zui3chun2 ‘lips’)

As shown above, 脸 (lian3 ‘face’), 眼 (yan3 ‘eyes’), 两臂 (liang3bi4 ‘arms’), 胸部/胸

(xiong1bu4/xiong1 ‘chest’), and 嘴唇 (zui3chun2 ‘lips’) are part of the human body. Cohesion in

this message is achieved by using five meronyms at the same time, but no superordinate.

Through using different parts of the human body, the meronyms make the description of Chief-

of-staff Lei and Lady Wu’s actions cohesive.

In the data, for this type of cohesion, the cases only show cohesion created by hyponymy. The

following three examples show cohesion through hyponymy.

In example [4:18], the micro-blog user talks about the gender imbalance of people working in the

government.

[4:18] (LWC, Message ID 3399982109657868)

1. 网友 建议 想 当官 的 男淫 做 变性 手术，

wang3you3 jian4yi4 xiang3 dang1guan1 de1 nan2yin2 zuo4 bian4xing4 shou3shu4.

Internet friend suggest want be leaders ASSC men take trans-gender surgery.

‘An Internet user suggested that men who want to be leaders should have transgender surgery.’

91  

 

2. 这 是 不错 的 主意。

zhe4 shi4 bu2cuo4 de1 zhu3yi1.

this be good ASSC idea.

‘This is a really good idea.’

3. 若干 年 后，

ruo4 gan2 nian2 hou4,

several year after,

‘After several years,’

4. 当 主席台 上 的 女 领导 坐 满 了 的 时候，

dang1 zhu3xi2tai2 shang4 de1 nv3 ling3dao3 zuo4 man3 le1 de1 shi4hou4,

when platform on ASSC female leader sit full PFV ASSC time,

‘when the platform is full of female leaders,’

5. 当官 的 机会 基本上 都 是 男淫 的 了，

dang1guan1 de1 ji1hui4 ji1ben3shang4 dou1 shi4 nan3yin2 de1 le1.

be leader ACCS opportunity basical all be men POSS CRS.

‘men would have the most opportunities to become leaders.’

6. 那 时候 又 轮到 女淫 做 变性 手术 了。

na4 shi3hou4 you4 lun2dao4 nv3yin2 zuo4 bian4xing4 shou3shou4 le1.

that time again turn women take trans-gender surgery CRS.

‘Then it would be women’s turn to have transgender surgery.’

92  

 

人 (ren2 ‘human’)

男淫 (nan2yin2 ‘men’) 女淫(nv3yin2 ‘women’)

In this example, 男淫 (nan2yin2, 男 ‘male’, 淫 ‘obscene’, 男淫 ‘men’) is a homophone of 男人

(nan2ren2 ‘men’). 女淫(nv3yin2, 女 ‘female’, 淫 ‘obscene’, 女淫 ‘women’) is a homophone of

女人 (nv3ren2 ‘women’ ). They are both hyponyms of 人 (ren2 ‘human’). The two Chinese

Internet homophones 男淫 and 女淫 link the theme to gender imbalance. Cohesion within this

message is achieved through the two hyponyms.

In example [4:19], the user talks about the change of their expectation on the baby’s gender. The

use of hyponyms makes the information of the gender change cohesive.

[4:19] (LWC, Message ID 3402514318079658)

1. 从 今天 起,

cong2 jin1tian1 qi3,

from today on,

‘From now on,’

2. 我 要 把 迎接 DD 的 心态 调试 成 迎接 MM 的，

wo3 yao4 ba3 ying2jie1 di4di de1 xin1tai4 tiao2shi4 cheng2 ying2jie1 mei4mei de1,

1stSG want BA welcome boy ASSC mentality change into welcome girl NOM,

‘I have to change my mentality of welcoming [a] boy to welcoming [a] girl.’

93  

 

3. 之前 的 心理 暗示 忒 强大 了，

zhi1qian3 de1 xin1li3 an4shi4 tui1 qiang2da4 le,

before ASSC mental suggestion too strong CRS,

‘[As] the previous mental suggestion is too strong,’

4. 要 一下 改 过来， 不 容易 啊

yao4 yi1xia4 gai3 guo4lai2, bu4 rong2yi4 a1.

want once change back, NEG easy PRT.

‘it is not easy to change [my mentality] quickly.’

宝宝 (bao3bao ‘baby’)

DD (di4di ‘boy’) MM (mei4mei ‘girl’)

As shown above, DD (di4di ‘boy’) is a homophone of 弟弟 (di4di ‘brother/young boy’) and MM

(mei4mei ‘girl’) is a homophone of 妹妹 (mei4mei ‘sister/young girl’). They are both hyponyms

of 宝宝 (bao3bao ‘baby’). The two hyponyms DD and MM serve as lexical cohesive devices to

express the user’s expectation of the baby’s gender has changed.

Example [4:20] is about the scene at a car show. Cohesion in this message is achieved through

the superordinate and its hyponyms.

94  

 

[4:20] (LWC, Message ID 3401766368879998)51

1. 不 来 一 发 吗 骚年！！！

bu4 lai3 yi4 fa1 ma1 sao1nian2!

NEG have a shot Q young person!

‘Why don’t [you] have a shot, young person!’

2. 我 想到 了 飞 天夜翔52 那 句 名言， 人 瘦 哔-----大。

wo3 xiang3dao4 le1 Fei1tian1ye4xiang2 na4 ju4 ming2yan2, ren2 shou4 bi4 da4.

1stSG recall PFV Feitianyexiang that CL famous saying, people slim then better.

‘I recalled the famous saying of Feitianyexiang, slim people look better.’

3. 现场 妹纸 的 脸上 都 洋溢 着 特殊 的 笑容，

xian4chang3 mei3zhi3 de1 lian3shang4 dou1 yang2yi4 zhe2 te4su4 de1 xiao4 rong2.

on the spot girls POSS face all fill with special ASSC smile.

‘All of the girls here have special smiles on their face.’

4. 拿着 单反 不 停 亮 闪光灯 的 汉纸 们 你们 赢 了

na2zhe2 dan1fan3 bu4 ting2 liang4 shan3guang1deng1 de1 han4zhi3 men2 ni3men2 ying2 le.

hold SLR53 NEG stop light flash ASSC boys 2ndPL win PFV.

‘The boys with non-stop flashing SLRs [cameras] have won.’

5. orz。

wo3 ren4 zai1.

I confess fall.

‘I confess.’                                                                                                                          51 This is an excerpt of the original message. 52飞天夜翔(Fei1tian1 Ye4xiang2) is a pen name of a Chinese online novelist. 53 SRLs is the short form of single-lens reflex.

95  

 

骚年 (sao1nian2 ‘young person’)

妹纸 (mei4zhi3 ‘girl’) 汉纸(han4zhi3 ‘boy’)

In this message, 骚年 (sao1nian2, 骚 ‘coquettish’, 年 ‘year’, 骚年‘young person’) is a

homophone of 少年 (shao4nian2 ‘young person’). 妹纸(mei4zhi3, 妹 ‘sister’, 纸 ‘paper’, 妹纸

‘girl’) is a homophone of 妹子 (mei4zi3 ‘girl’). 汉纸(han4zhi3, 汉 ‘sweat’, 纸 ‘paper’, 汉纸

‘boy’) is a homophone of 汉子 (han4zi3 ‘boy’). 妹纸 and 汉纸 are both hyponyms of the

superordinate 骚年. The homophonic superordinate and hyponyms create lexical cohesion within

the discourse and make the description of the scene cohesive.

4.3.4 Collocation

Collocation refers to the phenomenon that lexical items with a tendency to co-occur are used

together in discourse. These co-occurred words are related on a semantic level. Collocation can

appear in a same clause or different clauses. The lexical items that create collocation are

semantically associated. When one word occurs, other associated words will appear. For instance,

doctor, patient, and hospital, and rain and umbrella are lexical devices that help achieve

cohesion within discourse.

This type of cohesion is not very common in the data. The following example shows collocation

presented by a Chinese Internet homophone from the data.

96  

 

[4:21] (LWC, Message ID 3399857782088538)

1. 我 参与 了 新浪 微博 沃 3G54 随 想 随 行 活动,

wo3 can1yu3 le1 xin1lang4 wei1bo4 Wo4 3G sui3 xiang3 sui3 xing3 huo4dong4.

1ndSG participate PFV Sina Weibo Wo 3G whenever think whenever act activity.

‘I have participated in the Sina Weibo Wo 3G Mind with Action.’

2. 参与 证言人 话题 讨论, 获得                手机                    大奖                  抽奖                          机会，  

can1yu3 zheng4yan3ren2 hua4ti2 tao3lun4, huo4de2 shou3ji1 da4 jiang3 chou1jiang3 ji1 hui4.

join in witness topic discussion, get cell phone big prize draw chance.

‘Join in the discussion of witness, and you will have the chance to win a cell phone.’

3.  坚持            每天                  随身            随  心          织              围脖    就              有          更多                          好礼            等          你        拿！  

jian1chi2 mei3tian1 sui2shen1sui2xin1 zhi1 wei2bo2 jiu3 you3 geng4duo1 hao3li3 deng3 ni3 na2!

keep everyday carry satisfied knit micro-blog then have more good prize wait2ndPL take!

‘Keep “knitting” [your] micro-blog everyday and everywhere follow your inclinations, [and]

there will be more prizes waiting for you.’

围脖 (wei2bo2 ‘scarf’, here it refers to micro-blog’) is a homophone of 微博(wei1bo2 ‘Weibo,

micro-blog’). 织 (zhi1 ‘knitting’) in the context of this message, means to ‘write or use’ (micro-

blogs). The cohesion of collocation is achieved by 围脖 and 织. The two words 围脖 and 织 tend

to appear in the same lexical environment. The collocation of 围脖 and 织 make the theme of

promoting using Sina Weibo cohesive.

                                                                                                                         54沃 3G (wo4 san3 ji4 ‘Wo 3G’ ) is a type of 3G service provided by China Union (a Chinese telecommunication company).

97  

 

In this section, through the examination of Chinese Internet homophones in repetition,

synonymy/antonymy, hyponymy/meronymy, and collocation, I have shown that Chinese Internet

homophones can be used as lexical cohesive devices and enhance cohesion of the micro-blogs in

Chinese Internet discourse.

4.3 Summary

This chapter studied the cohesion function of Chinese Internet homophones by analyzing the

micro-blog data from the LWC. I have explored the function of Chinese Internet homophones

from two aspects: grammatical cohesion and lexical cohesion. When Chinese Internet

homophones serve as agents of reference, ellipsis, substitution, and conjunction, they function as

grammatical cohesive devices. Chinese Internet homophones can also function as lexical

cohesive devices when they are used in repetition, synonymy/antonymy, hyponymy/meronymy,

and collocation.

98  

 

Chapter 5 Conclusions

The use of new social media, such as micro-blogging, provides a platform for the acceleration of

the spread of Internet homophones. In previous chapters, I discussed the phonetic features and

morphological classification of Chinese Internet homophones, as well as their discourse function.

In this chapter I will summarize the findings of this thesis, and discuss the incentives for the

creation and use of Chinese Internet homophones. This will be followed by a discussion of the

limitations of this study and implications of the findings.

5.1 Summary of Findings

The research presented in this thesis combined a quantitative and qualitative analysis of Chinese

Internet homophones in the Leiden Weibo Corpus. In the data, 117 Chinese homophones were

identified. Among the top 10 most frequently used homophones in the data, the homophones are

generally used rather frequently compared to their base words. Based on their phonetic feature,

Chinese Internet homophones can be classified into two types: narrow-homophones which have

the same pronunciation as their base words, and broad-homophones which have a similar

pronunciation to their base words. In the dataset, 64 homophones (54.70%) are narrow-

homophones, while 53 homophones (45.30%) are broad-homophones. Narrow-homophones can

be further classified into homographic homophones (those that have the same spelling as their

base words or borrow ancient written forms) and heterographic homophones (those that have a

different spelling than their base words, including Chinese character heterographic homophones,

alphabet heterographic homophones, and numeric letter heterographic homophones). As for

99  

 

broad-homophones, this thesis examines Chinese character, alphabet, and numeric letter broad-

homophones. Compared with their base words, character broad-homophones tend to have the

same manner but different place of articulation as their base words. They may also display a

nasal [n] in homophones corresponding to a lateral [l] in the base words, or a lateral [l] or zero-

initials in homophones corresponding to a post-alveolar approximant r[ʐ]  in the base words. The

differences in finals of a syllable involve differing in medials, main vowels, and syllabic

terminals. The medials of homophones tend to be unrounded while that of the base words are

rounded. As for differences of the main vowels, the vowels of most homophones tend to be

higher and unrounded (e.g., a high front unrounded medial [i] in homophones corresponds to a

high front rounded medial [y] in the base words). Comparing the phoneme /e/ in the base words,

a high front unrounded [i] in homophones corresponds to a mid-central unrounded vowel [ə] in

the base words and a low central vowel [a] in homophones corresponds to a mid-high front

unrounded vowel [e] in the base words. In addition, the phonetic differences of homophones

involve the addition or deletion of vowels. As for the terminals, an alveolar nasal [n] and a velar

nasal [ŋ] can be interchanged sometimes. There is no fixed pattern for the tones of homophones

in the data compared with the tones of the base words. Chinese transliteration homophones

undergo minor sound changes compared to their pronunciations in the original language. Most of

the alphabet homophones retain the pronunciation of the Chinese characters from the base words.

The alphabet homophones are usually acronyms of the Pinyin of the base words, representing the

pronunciation of the base words. As for numeric letters, a single numeric letter can represent

different homophonic sounds, such as sounds from Mandarin, sounds from other dialects, and

sounds from other languages.

100  

 

Based on their morphological features, Chinese homophones can be classified into (Chinese)

character homophones, alphabet homophones, numeric letter homophones, and a blend of

characters, alphabets, and numeric letters (Zhou, 2008, p. 434). Among the 117 Chinese Internet

homophones identified in the data, there are 55 character homophones (47%), 46 alphabet

homophones (39.32%), 4 numeric letter homophones (3.42%), and 12 blending homophones

(10.26%). There are in total 75,592 occurrences of the 117 homophones in the dataset, among

which character homophones occurred the most frequently (55.08%), followed by alphabet

homophones (38.66%), numeric letter homophones (4.98%), and blending homophones (1.28%).

Character homophones and letter homophones are used much more frequently than numeric

letter homophones and blending homophones. Chinese character homophones are the most

frequently used type, which is consistent with Zhou (2008a)’s finding. As for Chinese character

homophones, they are classified into single-morphemic words (including monosyllabic single-

morphemic words and polysyllabic single-morphemic words) and multi-morphemic words

(including compounds, reduplications, and derivations). Single-morphemic words are used more

frequently than multi-morphemic words.  Alphabet homophones are derived from the Pinyin of

each Chinese base word. Alphabet homophones are written in Pinyin, Pinyin acronym, or Pinyin

with some variations. Acronym homophones are the most prominent in the category of alphabet

homophones. Numeric letter homophones consist of Arabic numbers. Numeric letters either

present the same or close sounds from Chinese, or the same or close sounds from other

languages, such as Japanese. Blending homophones are a mix of characters, alphabets, and

numeric letters. The frequency of blending homophones is the lowest among the four

morphological types of homophones. Some alphabets in blending homophones have similar

101  

 

pronunciations to those in English, unlike the letters in alphabet homophones, where they are

pronounced based on the Chinese sounds.

The qualitative analysis of homophones is based on the theoretical framework of systemic

functional grammar (Halliday & Hasan, 1976; Halliday, 1985; Halliday & Matthiessen, 2004). It

mainly studies the cohesion function of homophones in Chinese Internet discourse. Cohesion is

the connection within text that makes it a semantic unit (Halliday et al., 1976, p.4). In this study,

Hu et al. (2005)’s classification of cohesion was adopted, which includes grammatical cohesion

and lexical cohesion. Grammatical cohesion is further divided into four subcategories: reference,

ellipsis, substitution, and conjunction. The subcategories of lexical cohesion include repetition,

synonymy/antonymy, hyponymy/meronymy, and collocation. From the excerpts drawn from the

Leiden Weibo Corpus, homophones can function as grammatical and lexical cohesive devices in

Internet discourse. They make each message form a unified text, rather than a disconnected

sequence of clauses. However, the homophones that are used as conjunction (a subcategory of

grammatical cohesion), and collocation (a subcategory of lexical cohesion) devices are not very

common in the data.

5.2 Discussion

The emergence of Chinese Internet homophones is inseparable from their own characteristics

and the linguistic, social, and cultural environment in which they are used. The following section

will briefly discuss the reasons for the creation of Chinese homophones. They are phonology of

Mandarin Chinese, the use of typing method for Mandarin, the influence of Chinese dialects, and

102  

 

the Internet censorship of the Chinese government. These factors facilitate the creation and use

of Chinese homophones in computer-mediated communication.

5.2.1 Phonology of Mandarin Chinese

Mandarin Chinese is a tonal language, with 21 initials, 39 finals, and 4 basic tones in its

phonological system. Given the small set of initials and finals, the number of syllables that can

be formed is limited. According to the Mandarin phonological rules and conventions, there can

be only 398 sets of initial-final combinations, and approximately 1,200 distinct tonal syllabic sets

(Suen, 1979). However, there are 4,574 Chinese characters in common usage (Modern Chinese

Frequency Dictionary, 1986). As a result, there exists a great amount of homophonic words with

the same or close sound as their base words.

5.2.2 Typing Method

Many people believe that Chinese, with its character-based orthography, could not serve for

electronic use. But by 2006, it had become the second most commonly used language on the web

(Gottlieb, 2009). Most Chinese Internet users rely on the Pinyin typing method. First, they type

out the Pinyin of a word through the keyboard, and then they select from a pop-up word list that

contains words with the same initial and final. Finally, they choose the desired word, which is

displayed on the screen. Because the words in the typing method system are arranged based on

the frequency of common lexicon. In fast-paced Internet communication, Internet users may

choose to use homophonic words and the most concise script to express their intended meaning

103  

 

and to speed up the typing and the exchange of information. Some Internet users use

homophonic puns, disregarding the accuracy of their typing and the language norm. This is

because, for some words, their homophonic puns are faster and more convenient to type. For

example, if Internet users want to type the word 版主 ban3zhu3‘board moderator’ by Pinyin,

the word 斑竹 ban1zhu2 ‘mottled bamboo’ comes first on the list. To save time, more and more

Internet users use 斑竹 instead of 版主, and gradually 斑竹 has become a homophonic pun of 版

主. This is also because that the Internet word 版主 ranked relative low in the frequency of the

common lexicon when this word was stared to use on the Internet.

For most alphabet homophones, their emergence is also related to the Pinyin typing method.

Instead of scrolling down the word list and finding the corresponding word, Internet users press

the Capital Lock key on the keyboard, and type the full pinyin form or acronym, such as Ding

ding3 ‘to support or to push up the ranking of a posting’ (base word 顶 ding3 ‘to support’) and

ZZ zhuan3zai3 ‘to forward others’ posting’ (base word 转载 ‘to forward’).

5.2.3 The Influence of Chinese Dialects

In Mainland China, there are 7 main dialect groups distinguished by their phonological and

phonetic differences, but they share the same writing system (i.e. using simplified Chinese

characters). With the prevalence of local on-line chat rooms and local BBS, and the popularity of

youth culture, various spoken words in Chinese dialects have acquired written forms, with

phonetic borrowing from Chinese characters. These new words have become popular among

104  

 

other Internet users across the country. For example, the use of 虾米 xia1mi3 ‘little shrimp’ as a

homophone of 什么 shen3me ‘what’ started from Southern Fujian dialect speakers. Later it was

gradually adopted and frequently used by other young Internet users nationwide.

5.2.4 Internet Censorship by the Chinese Government

Mainland China is considered to have the most pervasive and sophisticated regime of Internet

filtering and information control in the world (John, Palfrey, Jonathan, & Zittrain, 2011).

Originally, the censorship of the Internet carried out by the Chinese government was to “purify”

the language environment (i.e., to filter obscene and erotic words). Since the early 2000’s, more

and more “politically sensitive” words have been censored in order to maintain social stability.

For example, 党 dang3 ‘party’ and 政府 zheng4fu3 ‘government,’ which are frequently used in

comments that are regarded as threatening to social stability, have been censored. As a result, the

Internet censorship restricts Internet users’ online postings. As an alternative, Internet users

started to use Chinese character homophones and acronym alphabet homophones in order to pass

the filter and publish their words online. For example, 河蟹 he2xie4 ‘river crab’ is used as a

homophone of 和谐 he2xie3 ‘harmonious’ (a political concept of the government) and GCD

gong4chan3dang3 ‘the Communist Party’ is used as a homophone of 共产党 gong4chan3dang3

‘the Communist Party’.

In addition, other factors, such as the influence of foreign languages and multilingual youth

culture, also facilitate the creation and use of Chinese homophones on the Internet.

105  

 

5.3 Limitations of the Study

Internet language, known as a type of social dialect, is changing along with social development.

The lexicon is one of the most sensitive parts in Internet discourse, and it is affected by

technology, politics, society, culture, and the language environment. Internet users’ gender, age,

and educational background also play an important role in computer-mediated communication.

The present study has several limitations. The first drawback is its word list of Chinese Internet

homophones. As constrained by the large size of the corpus, the initial word list was constructed

from the Green Paper on Language Situation in China (2006-2013), Chinese Internet language

dictionaries, and Sina Weibo Niandu Pandian (Sina Weibo Annual Report, 2011-2012). This

may have caused certain homophones to be excluded. Second, the calculation of the total

occurrences of each homophone found within the corpus is done by the software CasualConc.

Because the data is too large (i.e. 18,791,530 words ), only part of the data that contained target

homophones was checked manually. As a result, there may be some minor errors in the

frequency statistics. Finally, as this study started in the spring of 2012, the data were collected in

January 2012. Therefore any homophones that came into use after January of 2012, along with

any new features, are not included in this study.

5.4 Implications of the Study

This study contributes to the fields of Chinese linguistics, systemic functional linguistics,

language variation and planning, and second language teaching.

106  

 

Some of the more frequently used homophones have already entered the lexicon of computer-

mediated communication. With the interaction of computer-mediated communication and daily

communication, some words may become common words within the Chinese vocabulary system,

while other words may disappear. From the phonetic and morphological studies of Chinese

Internet homophones, we can see the development of a trend within Chinese Internet language,

i.e. more than half of the homophones retain the same pronunciation as their base words, and

Chinese character homophones are the most frequently used type. It is also interesting to see the

changes and variations of Chinese Internet language, which can be used for future language

planning.

From a systemic functional grammar perspective, homophones can be used as lexical and

grammatical cohesive devices. Homophones serve as cohesive devices to make the discourse a

complete semantic unit. This paper shows that Hu et al. (2005)’s model can be applied to

different styles of written texts, including Internet discourse. The study of the cohesion function

of Chinese Internet homophones allows us to see the semantic meaning of whole text and the

connections between textual units within the discourse. The analysis of the cohesion function

also contributes to our understanding of the relationship between discourse and context. The

findings of the cohesion analysis of Internet homophones can also be applied to second language

teaching and learning to help Chinese language learners to understand Chinese Internet language

and to communicate with Chinese native speakers more efficiently.

107  

 

Reference “Annual Inventory Check of 2011 Sina Mirco-blog.” (n.d.). In Sina Tech. Retrieved from http://tech.sina.com.cn/z/2011pandian/index.shtml “Annual Inventory Check of 2010 Sina Mirco-blog.” (n.d.). In Sina Tech. Retrieved from http://tech.sina.com.cn/i/2010-12-29/07505040121.shtml Bo, J. L. 2007. “Homophobic Puns and the Purpose of Expression.” Science & Technology

Information 30: 195 & 205. Brown, G. & Yule, G. 1983. Discourse Analysis. Cambridge: Cambridge University Press. Butler, C. 1985. Statistics in Linguistics. Oxford: Basil Blackwell. Campell, K. S. 1995. Continuity and Cohesion: Theoretical Foundation of Document Design.

New Jersey: Lawrence Erlbaum Association. Cao, W. 2004. The Study of Modern Chinese Lexicons. Beijing: Peking University Press. Chen, D. W. 2007. “Analysis on Discourse Cohesion in Chinese Cross Talk.” M.A. Dissertation,

South China Normal University. Chen, R. 2004. Contemporary Chinese Rhetoric. Beijing: Peking University Press. Chen, P. 2011. “A Comparative Study of Xieyin in Chinese and English from the Perspective of

Cognition.” M.A. dissertation, Central China Normal University. Collot, M. & Belmore, N. 1996. “Electronic Language: A New Variety of English.” Pp. 13-28 in

Computer-Mediated Communication: Linguistic, Social, and Cross-culture Perspectives edited by Herring S. John Benjamins Publishing Company.

Crystal, D. 2006. Language and the Internet. Cambridge: Cambridge University Press. Enkvist, N.E. 1978. “Coherence, Pseudo-Coherence, and Non-coherence.” Pp. 109-128 in Report

on Text Linguistics: Cohesion and Semantics edited by Ostman, J. O. Turku-Abo, Finland: Research Institute of the Abo Academy Foundation.

Fang, H. F. & Li, Y. B. 2011. “Literature Review on Text Cohesion and Coherence Theory in

China.” Data of Culture and Education 5: 37-39.

108  

 

Fang, Y. 2005. “Systemic Functional Grammar and Discourse Analysis.” Foreign Language Education 5: 1-5.

Fang, Z. 2005. “On the Internet Vocabulary Formation.” Rhetoric Learning 6: 39.

Feng, J. (ed.). 2012. New Internet Dictionary. Beijing: New World Press. Fong, M. 2000. “ ‘Luck Talk ’ in Celebrating the Chinese New Year.” Journal of Pragmatics 32:

219-237. Fu, H.Q. 1985. Modern Chinese Lexicon. Beijing: Peking University Press. Ge, B. Y. 1985. On Chinese Word-formation and Morphology. Journal of Literature, History and Philosophy 4: 28-33. Ge, B.Y.2001. Modern Chinese Lexicology. Shandong: Shandong People’s Publishing House. Georgakopoulou A. 2011. “Computer-mediated Communication.” Pp. 93-106 in Handbook of

Pragmatics Hightlights 9: Pragmatics in Practice edited by Fried, M., Jan-Ola Östman, & Verschueren, J. John Benjamins Publishing Company.

Gottlieb, N. 2009. “Language on the Internet in Japan.” Pp. 65-78 in Internationalizing Internet

Studies: Beyond Anglophone Paradigms edited by Goggin, G. & McLelland, M. New York & London: Routledge.

Gu, W, Pan, W. Y., & Liu, Y. H. 2001. Practical Modern Chinese Grammar. Beijing: The

Commercial Press. Guo, H.J. 2002. “English Loan Words in Contemporary Chinese and Their Impact on Chinese

Grammar: the Period from 1980s to the Present.” Journal of PLA University of Foreign Language 05: 1-4.

Guo, Z.R. 2001. “On Homophonic Rhetoric.” Chinese Character Culture 4: 50-52. Guo, Y. K. 1999. “The Changes of Homophonic Pun.” Rhetoric Learning 5: 6-7. Gutwinski, W. 1976. Cohesion in Literary Texts: A Study of some Grammatical and Lexical

Features of English Discourse. The Hague: Mouton Linguarum Series Minor 204. Halliday, M. A. K. 1962. “Descriptive Linguistics in Literary Studies.” Pp. 26-39 in English

Studies Today, 3rd Series edited by Duthie, G. I. Edinburgh University Press. Halliday, M.A.K. 1964. “The Linguistic Study of Literary Texts.” Pp. 302-307 in Proceedings of

the Ninth International Congress of Linguists edited by Lunt, H. G. The Hague: Mouton. [reprinted in Chatman , S. & Levin, S. R.1967. Essays on the Language of Literature. Boston: Houghton Mifflin. 217-223.]

109  

 

Halliday, M. A. K. 1968. “Notes on Transitivity and Theme in English, Part 3.” Journal of Linguistics 4:179-215.

Halliday, M. A. K. 1970. “Language Structure and Language Function.” Pp. 140-165 in New

Horizons in Linguistics, edited by Lyons, J. Harmondsworth: Penguin. Halliday, M.A.K. & Hasan, R.1976. Cohesion in English. London: Longman. Halliday, M. A. K. 1985. An Introduction to Functional Grammar. London: Arnold. Halliday, M. A. K. & Hasan, R. 1985. Language, Context and Text: Aspects of Language in a

Social Semiotic Perspective. Victoria, Australia: Deakin University Press. Halliday, M.A. K. 1994. An Introduction to Functional Grammar (2nd Edition). London:

Arnold. Halliday, M.A. K. & Matthiessen, C. 2004. An Introduction to Functional Grammar. New York:

Oxford University Press. Hartmann, R.R.K., & Stork, F.C.1972. Dictionary of Language and Linguistics. London:

Applied Science. Herring, S. 1996. “Introduction.” Pp. 1-12 in Computer-Mediated Communication: Linguistic,

Social, and Cross-culture Perspectives edited by Herring, S. John Benjamins Publishing Company.

Hoey, M. 1991 Patterns of Lexis in Texts. Oxford: Oxford University Press. Hu, G. M. 2006. “The Study of Chinese Harmonic Phenomenon”, M.A. dissertation, Changchun

University of Science and Technology. Hu, Z. L. 1984. “Differences in Mode.” Journal of Pragmatics 8: 595-606. Hu, Z. L. 1993. “The Cohesive Function of the Phonological System in Producing English

Texts.” Foreign Language Teaching and Research 2: 1-8. Hu, Z. L. 1994. Discourse Cohesion and Coherence. Shanghai: Shanghai Foreign Language

Education Press. Hu, Z. L. 1996. “Thought on Multiple Levels of Textual Cohesive Theory.” Journal of Foreign

Languages1: 1-8. Hu, Z.L., Zhu, Y. S., Zhang, D. L., & Li, Z.Y. 2005. Introduction to System-Functional

Linguistics. Beijing: Peking University Press.

110  

 

Hu, Y.S. 1992. Modern Chinese. Hong Kong: Joint Publishing. Huang, B. R., & Liao, X. D. (ed.) 2008. Modern Chinese. (3rd Edition) Beijing: Higher

Education Press. Huang, G.W. 1988. Essentials of Text Analysis. Hunan, China: Hunan Education Press. Huang, G. W. 2000. “Systemic Functional Linguistics: Forty Years on.” Foreign Language

Teaching and Research 1: 15-21. Huang, G.W. 2001. Theory and Practice of Discourse Analysis. Shanghai Foreign Language

Education Press. Huang, G. W. 2002. “Hallidayan Linguistics in China.” World Englishes 21(2): 281-290. Huang, G. W. 2011. “Systemic Functional Grammar Research in China: Its Development and

Future.” Pp. 42-78 in A General Introduction to Systemic Functional Linguistics edited by Huang, G. W. & Xin, Z. Y. Beijing: Foreign Language Teaching and Research Press.

Institute of Linguistic Studies, Beijing Language and Culture University (ed.). Modern Chinese

Frequency Dictionary. 1986. Beijing Language and Culture University Press. Jiang, Y. 2003. “Summary on the Network Language Derived Principle.” Rhetorical learning 2:

32. Joel B. 2004. “Second Language Cyber Rhetoric: a Study of Chinese L2 Writers in an Online

Usenet Group.” Language Learning & Technology 8(3): 66-82. Li, C. L. 2009. “On Discourse Mechanism of Homophonic SMS.” M.A. Dissertation, Hunan

Normal University. Li, C. Y. 2009. “The Formation and Features of Internet Catchphrases”. Guangxi Social

Sciences 2: 114-116. Li, L. X. 2007. “Network Language Research.” M.A Dissertation, Shaanxi Normal University. Li, M. 2004. “The Types, Characters and Criterion of Words in the Internet Language.” Journal

of Women’s Academy at Shandong 3: 48-50. Li, J. 2003. “‘Cyber language’ and Its Space.” Beijing Morning Post 19 (5). Li, E. S.2007. A Systemic Functional Grammar of Chinese. London: Continuum International

Publishing Group.

111  

 

Liao, Y. 2004. Linguistic Study on News Report: Cohesion and Coherence of News Text. PHD Dissertation, Hunan Normal University.

Lin, J. C. 1986. “Exploration of Lexical Devices in English Texts.” Journal of Foreign

Languages 5: 20-26. Lin, G. 2002. “The Types and Characteristics of the Network Term.” Rhetoric Learning1: 26. Liu, C.M. 2002. “‘Network Token’ Group Is Reasonable, According to the Analysis.” Language

Teaching and Research 6: 31-35. Liu, H. R. 2010. “Overview of Chinese Internet Language Research from 2000 to 2009.” M.A.

Dissertation, Northeast Normal University. Liu, J. 2011. “Deviant Wring and Youth Identity: Representation of Dialects with Chinese

Characters on the Internet.” Chinese Language and Discourse 2: 58-79. Liu, Q. Y. 2004. “Cohesion and Coherence in Text Translation.” Shangdong Foreign Languages

Journal 3: 5-9. Liu, X. M. 2006. “Homophonic Puns in Internet Language.” Journal of Language and Literature

Studies 9: 129-131. Lu, H.2003. “On Research of Voice Variation Phenomenon of Internet Language.” Rhetoric

Learning 1: 38. Luo, X. M. & Liu, Q. R. 2003. “The Coherent Function of Lexical Chains in Discourse.”

Foreign Language Education 2: 7-10. Lü, S. X. 1984. Papers in Chinese Syntax. Beijing: The Commercial Press. Martin, J. 1992. English Text: System and Structure. Amsterdam: John Benjamins. Martin, J. 2001. “Cohesion and Texture.” Pp. 35-53 in The Handbook of Discourse Analysis

edited by Schiffrin, D., Tannen, D., & Hamilton, H. E. Malden, Mass: Blackwell Publishers.

McCarthy, M. & Carter, R. 1994. Language as Discourse. London and New York: Longman. McDonald, E.1992. “Outline of a Functional Grammar of Chinese for Teaching Purposes.”

Language Sciences 14 (4): 435-458. Miao, X. W. 1998. “The Relationship between Cohesion and Coherence.” Journal of Foreign

Languages 4: 44-49.

112  

 

Morgan, J. L. & Sellner, B.M. 1980. “Discourse and Linguistic Theory.” Pp. 165-200 in Theoretical Issues in Reading Comprehension edited by Spiro, R. J., Bruce, B.C. & Brewer, F.W. New Jersey: Lawrence Erlbaum.

Nigel, I. 2010. “China in Cyberspace.” Survival: Global Politics and Strategy 52(4): 55-66. Ning, S. W. 1975. “ ‘Jiang zhenshi yinqu’ ‘yong jiayu cunyan’ [to Hide Truth and Use Rumor]”.

Journal of Nanjing Normal University (Social Science Edition) 3: 70-75. Oxford Language Dictionaries Online, http://www.oxfordlanguagedictionaries.com Packard, J.L. 2000. The Morphology of Chinese: A Linguistic and Cognitive Approach.

Cambridge University Press. Pan, G.W., Ye, B.Q., & Han, Y.2004. Research on Chinese Morphology. Shanghai: East China

Normal University Press. Parivash, M. & Bahram, L. 2011. “Introducing Categorization of Homophonous Words in

English Language.” Procedia-Social and Behavioral Sciences 31: 602-606. Parsons, G. 1990. Cohesion and Coherence: Scientific Texts: A comparative Study. Department

of English Studies, University of Nottingham. Partington, A. 2009. “A Linguistic Account of Wordplay: the Lexical Grammar of Punning.”

Journal of Pragmatics 4: 1794-1809. Paula L. R. 2007. “Teaching L2 Vocabulary through SMS Language: Some Didactic Guidelines.”

Estudios de linguistic ingleca aplicada 7: 165-188. Platt, J., Richards, J., & Webber, H. 1985. Longman Dictionary of Applied Linguistics. Great

Britain: The Richard Clay (The Chaucer Press) Ltd. Pi, H.M. 1992. “The Features and Significance of Bi-syllablization in Chinese.” Research in

Ancient Chinese Language 01: 56-59. Pollack, J. 2011. The Pun Also Rises. New York: Gotham Press. Qi, X. J. 2002. “On the Semantics of Homophones in Internet Language.” Studies of Rhetoric 3:

21. Redfern, W.1984. Puns. London: Blackwell. Ronald J., P. Edibert, J. Rafal, & Z. Rohozinski 2011. Access Contested: Security, Identity, and

Resistance in Asian Cyberspace. Massachusetts: The MIT press.

113  

 

Shi, C.H. 2009. “The Language Value and Linguistic Value of Web Language.” Proceedings from the 6th Academic Forum on Chinese Language Application. Retrieved from http://www.dwhyyjzx.com/www/news.asp?id=22&newsid=346.

Shi, Y. L. & Sun, Z. X. 2011. “A Comparative Study on the Cohesion in Treasury Island and Its

Chinese Version.” Writer Magazine 3: 193-194. Song, Y.Z.1992. Basic Knowledge of Modern Chinese. Beijing: Language & Culture Press. Suen , C. Y. 1979. Computational Analysis of Mandarin. Basel, Switzerland: Birkhauser. Sun, C.C. 2011. “From the Rapid Population of beiju to See the Development Trend of Internet

Homophones.” M.A.Dissertation, Jilin University. Sun, C.F. 2006. Chinese: A Linguistic Introduction. Cambridge University Press. Sun, Y. 1995. “Commentary on Hu Zhuanglin’s Discourse Cohesion and Coherence.” Journal of

Foreign Languages 4: 78-79. Sun, Y. 1997. “On the Origin, Nature and Relation of Cohesion and Coherence.” Journal of

Foreign Languages 1: 31-35. Sun, Y. 2011. “Research on Systemic Functional Grammar and Russian Textual Coherence.”

PHD Dissertation, Heilongjiang University.

Tang, Q. Y. 2005. “Exploration of Cohesion in Medical Discourse: Discourse Cohesion Analysis on an Eye Drop Advertisement.” Journal of Jinggagnshan Medical College 2 (4): 18-19.

Thompson, G. 2004. Introducing Functional Grammar. London: Arnold.

Thorne, S. 2008. “Computer-Mediated Communication.” Pp. 1-12 in Encyclopedia of Language and Education, 2nd Edition, Volume 4: Second and Foreign Language Education edited by Deusen-Scholl, V & N. Hornberger (eds). Spring Science+ Busniess Media LLC.

Wang, C. L. 2000. “On the Anaphoric Function of The Personal Pronoun ta.” Studies of the

Chinese Language 3: 228-237.

Wang, H. Y. 2010. “On Grammatical and Phonetic Cohesion of Antithetical Couplets.” Literature and Arts 4: 15-16.

Wang, J. 2005. “A Study on the Cohesion Means Based on Grammar and Sound”. M.A. Dissertation, Nanchang University.

Wang, P. (2000). “Objective Basis for Chinese homonym.” Rhetoric Learning 2: 8-9.

114  

 

Wang, Z. Q. 2002. “Another Discussion on Cohesion, Coherence and Correlation.” Foreign Language Research 2: 28-31.

Wang, Z. Y. 1981. “Review of Halliday’s ‘Grammatical Categories in Modern Chinese.” Linguistic abroad 2:55-62.

Wei, N.S. 2011. “Detailed Analysis of Chinese Character Homophones and Homophones with

Borrowed Base words In the Internet.” Modern Chinese 08: 61-62. Widdowson, H. C. 1978. Teaching Language as Communication. Oxford: Oxford University

Press. Wu, D. Y. 2001. “On Englishization of Chinese Morphology.” Contemporary Linguistics 2: 81-

89. Wu, W. Z. & Tian, X. L. 2000. Clusters of Chinese Sentences. Beijing: The Commercial Press. Xiang, Z. 2012. “The Research of Verbal Communication Characteristics in MicroBlog.” MA

Dissertation, Jilin University, China. Xu, S. H. 1982. On Theme and Rheme. Foreign Language Teaching and Research 2: 1-8. Xu, T.J. 1993. Language. Changchun: Northeast Normal University Press. Xu, J. 2004. “Cohesion, Texture, and Coherence.” PHD Dissertation, Fudan University. Xu, Y. C. 1996. “Comparative Studies on Cohesive Devices in Chinese and English.”

Shangdong Foreign Language Journal 4: 1-6.

Xu. Y. C. 2003 a. “Comparative Analysis on English Essay Cohesive Devices and Errors.” Foreign Language Teaching Abroad 3: 54-60.

Xu. Y. C. 2003b. “Textual Cohesion and Mental Coherence.” Shangdong Foreign Language Journal 3: 12-16.

Yan, B. 2005. “On Partial Tone Use in Network Language.” Journal of Changdu University of Technology Social 13(3): 32-34.

Yan, X. 2010. “Cohesion Studies in the Past 30 Years: Development, Application and Chaos.”

The International Journal - Language Society and Culture 31: 139-147. Ye, H. 2005. “Analysis of the Impact of Internet Language to Chinese and Its Future.” August

12, http://www.hpe.sh.cn.

115  

 

You, Y.X. 2012. “Characteristics of Neologism and Its Cognitive Motivations: A Study of Chinese Neologisms from 2006 to 2009.” PhD Dissertation, Shanghai International Studies University.

Yu, G. Y. (ed.) 2001. Chinese Internet Language Dictionary. Beijing: China Economic

Publishing House. Yu, Y. 2009. “On the Internet Word jiong.” Studies of Rhetoric 1: 93-95. Yukiko N.2003. “Linguistic Innovations and Interactional Features of Casual On-line

Communication in Japanese.” Journal of Computer-Mediated Communication 9 (1), http://jcmc.indiana.edu/vol9/issue1/nishimura.html

Zhang, B. J. & Fang, M. 1996. Functional Studies of Chinese Grammar. Beijing: Commercial

Press. Zhang, D.L. 1999. “A Survey of Studies on Text Coherence.” Journal of Foreign Languages 6:

24-31. Zhang, D. L. 2001a. Cohesive Force and Text Coherence. Foreign Languages and Their

Teaching 1: 9-15. Zhang, D. L. 2001b. “On Cohesion.” Journal of Foreign Languages 2: 23-28. Zhang, D. L. 2001c. “On the Principles of Internal Cohesion of Text.” Journal of PLA University

of Foreign Languages 11: 26-32.

Zhang, D. L. 2002. “Cohesion and Style: The Stylistic Features of Reference and Lexical Cohesion.” Foreign Language and Their Teaching 10: 1-7.

Zhang, D. L. & Liu, R. S. 2003. Progress in Studies on Text Cohesion and Coherence. Shanghai: Shanghai Foreign Language Education Press.

Zhang, D. L. 2004. “New Developments in Systemic Functional Linguistics.” Contemporary Linguistics 6: 57-65.

Zhang, D. L. 2005. “Form and Meaning in Text Cohesion.” Journal of Foreign Languages 5: 32-38.

Zhang, D. L. 2006. “Systemic Functional Linguistics in China.” Foreign Languages in China 02: 27-32.

Zhang, Y. H. 2007. “On Grammatical Features of Internet Lexicon.” Studies of the Chinese Language 6: 531-535.

Zhang, Y.Y. 1982. Brief Introduction of Lexicology. Wuhan: Huazhong University of Science and Technology Press.

116  

 

Zhao, H. L. 2005. “On the Rhetoric of Internet language.” Applied Linguistics 9: 123-125.

Zhao, J. M. 1987. “Homophony and Culture.” Language Teaching and Linguistic Studies 1: 40-56.

Zheng, Q. J. 2007. “A Study of Intertextual Cell Phone Messages and Their Texture.”  Language Teaching and Linguistic Studies 05: 82-89.

Zhou, C. L. 2005. “On Homophony’s Role, Word-Formation, and Rhetoric Effect”. Economic

and Social Development, 2: 4-8. Zhou, H. Y. 2006. “The Application of Cohesion in Text Functional Analysis.” Foreign

Languages and Their Teaching 10: 22-24. Zhou, Q.S. (ed.) 2006 a. Language Situation in China: 2005 Part 1. Beijing: The Commercial

Press. Zhou, Q.S. (ed.) 2006 b. Language Situation in China: 2005 Part 2. Beijing: The Commercial

Press. Zhou, Q.S. (ed.) 2007 a. Language Situation in China: 2006 Part 1. Beijing: The Commercial

Press. Zhou, Q.S. (ed.) 2007 b. Language Situation in China: 2006 Part 2. Beijing: The Commercial

Press. Zhou, Q.S. (ed.) 2008 a. Language Situation in China: 2007 Part 1. Beijing: The Commercial

Press. Zhou, Q.S. (ed.) 2008 b. Language Situation in China: 2007 Part 2. Beijing: The Commercial

Press. Zhou, Q.S. (ed.) 2009 a. Language Situation in China: 2008 Part 1. Beijing: The Commercial

Press. Zhou, Q.S. (ed.) 2009 b. Language Situation in China: 2008 Part 2. Beijing: The Commercial

Press. Zhou, Q.S. (ed.) 2010 a. Language Situation in China: 2009 Part 1. Beijing: The Commercial

Press. Zhou, Q.S. (ed.) 2010 b. Language Situation in China: 2009 Part 2. Beijing: The Commercial

Press.

117  

 

Zhou, Q.S. (ed.) 2011. Language Situation in China: 2010. Beijing: The Commercial Press. Zhou, Q.S., & Hou, M. (ed.) 2012. Language Situation in China: 2011. Beijing: The

Commercial Press. Zhu, D.X. 1982. Lecture Notes of Grammar. Beijing: The Commercial Press. Zhu, Y. S. 1995. “The Development and Improvement of Cohesion Theory.” Journal of Foreign

Languages 3: 36-41.

Zhu, Y. S., Yan, S. Q., & Miao, X. W. 2001. Contrastive Study of Cohesion in English and Chinese. Shanghai Foreign Language Education Press.

Zhu, Y. S. 2002. “Semantic Indirectness in Chinese Compounds and the Discrimination between

Compounds and Phrases.” Chinese Teaching in the World 05: 50-59.

Zuo, Y.1995. “Comparing Cohesive Device in Chinese and English.” Foreign Language

Teaching and Research 3: 37-42.

118  

 

Appendix

Abbreviations used for the glosses in the Chinese examples:

ADV an adverbial maker (de)

ASSC associative (de, zhi)

BA a pretransitive maker (ba)

CL classifier

CRS currently relevant state (le)

CSC complex stative construction (de)

NEG negatives (bu, mei, bie)

NOM nominalizer

PFV perfective aspect (le)

PL plural personal pronoun

POSS possessive (-de)

PRT particle

Q question marker (ma)

SG singular personal pronoun